1 /*
2 * parserInternals.c : Internal routines (and obsolete ones) needed for the
3 * XML and HTML parsers.
4 *
5 * See Copyright for the status of this software.
6 *
7 * daniel@veillard.com
8 */
9
10 #define IN_LIBXML
11 #include "libxml.h"
12
13 #if defined(_WIN32)
14 #define XML_DIR_SEP '\\'
15 #else
16 #define XML_DIR_SEP '/'
17 #endif
18
19 #include <string.h>
20 #include <ctype.h>
21 #include <stdlib.h>
22
23 #include <libxml/xmlmemory.h>
24 #include <libxml/tree.h>
25 #include <libxml/parser.h>
26 #include <libxml/parserInternals.h>
27 #include <libxml/entities.h>
28 #include <libxml/xmlerror.h>
29 #include <libxml/encoding.h>
30 #include <libxml/xmlIO.h>
31 #include <libxml/uri.h>
32 #include <libxml/dict.h>
33 #include <libxml/xmlsave.h>
34 #ifdef LIBXML_CATALOG_ENABLED
35 #include <libxml/catalog.h>
36 #endif
37 #include <libxml/chvalid.h>
38
39 #define CUR(ctxt) ctxt->input->cur
40 #define END(ctxt) ctxt->input->end
41
42 #include "private/buf.h"
43 #include "private/enc.h"
44 #include "private/error.h"
45 #include "private/io.h"
46 #include "private/parser.h"
47
48 /*
49 * XML_MAX_AMPLIFICATION_DEFAULT is the default maximum allowed amplification
50 * factor of serialized output after entity expansion.
51 */
52 #define XML_MAX_AMPLIFICATION_DEFAULT 5
53
54 /*
55 * Various global defaults for parsing
56 */
57
58 /**
59 * xmlCheckVersion:
60 * @version: the include version number
61 *
62 * check the compiled lib version against the include one.
63 * This can warn or immediately kill the application
64 */
65 void
xmlCheckVersion(int version)66 xmlCheckVersion(int version) {
67 int myversion = LIBXML_VERSION;
68
69 xmlInitParser();
70
71 if ((myversion / 10000) != (version / 10000)) {
72 xmlGenericError(xmlGenericErrorContext,
73 "Fatal: program compiled against libxml %d using libxml %d\n",
74 (version / 10000), (myversion / 10000));
75 fprintf(stderr,
76 "Fatal: program compiled against libxml %d using libxml %d\n",
77 (version / 10000), (myversion / 10000));
78 }
79 if ((myversion / 100) < (version / 100)) {
80 xmlGenericError(xmlGenericErrorContext,
81 "Warning: program compiled against libxml %d using older %d\n",
82 (version / 100), (myversion / 100));
83 }
84 }
85
86
87 /************************************************************************
88 * *
89 * Some factorized error routines *
90 * *
91 ************************************************************************/
92
93
94 /**
95 * xmlErrMemory:
96 * @ctxt: an XML parser context
97 * @extra: extra information
98 *
99 * Handle a redefinition of attribute error
100 */
101 void
xmlErrMemory(xmlParserCtxtPtr ctxt,const char * extra)102 xmlErrMemory(xmlParserCtxtPtr ctxt, const char *extra)
103 {
104 if ((ctxt != NULL) && (ctxt->disableSAX != 0) &&
105 (ctxt->instate == XML_PARSER_EOF))
106 return;
107 if (ctxt != NULL) {
108 ctxt->errNo = XML_ERR_NO_MEMORY;
109 ctxt->instate = XML_PARSER_EOF;
110 ctxt->disableSAX = 1;
111 }
112 if (extra)
113 __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER,
114 XML_ERR_NO_MEMORY, XML_ERR_FATAL, NULL, 0, extra,
115 NULL, NULL, 0, 0,
116 "Memory allocation failed : %s\n", extra);
117 else
118 __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER,
119 XML_ERR_NO_MEMORY, XML_ERR_FATAL, NULL, 0, NULL,
120 NULL, NULL, 0, 0, "Memory allocation failed\n");
121 }
122
123 /**
124 * __xmlErrEncoding:
125 * @ctxt: an XML parser context
126 * @xmlerr: the error number
127 * @msg: the error message
128 * @str1: an string info
129 * @str2: an string info
130 *
131 * Handle an encoding error
132 */
133 void
__xmlErrEncoding(xmlParserCtxtPtr ctxt,xmlParserErrors xmlerr,const char * msg,const xmlChar * str1,const xmlChar * str2)134 __xmlErrEncoding(xmlParserCtxtPtr ctxt, xmlParserErrors xmlerr,
135 const char *msg, const xmlChar * str1, const xmlChar * str2)
136 {
137 if ((ctxt != NULL) && (ctxt->disableSAX != 0) &&
138 (ctxt->instate == XML_PARSER_EOF))
139 return;
140 if (ctxt != NULL)
141 ctxt->errNo = xmlerr;
142 __xmlRaiseError(NULL, NULL, NULL,
143 ctxt, NULL, XML_FROM_PARSER, xmlerr, XML_ERR_FATAL,
144 NULL, 0, (const char *) str1, (const char *) str2,
145 NULL, 0, 0, msg, str1, str2);
146 if (ctxt != NULL) {
147 ctxt->wellFormed = 0;
148 if (ctxt->recovery == 0)
149 ctxt->disableSAX = 1;
150 }
151 }
152
153 /**
154 * xmlErrInternal:
155 * @ctxt: an XML parser context
156 * @msg: the error message
157 * @str: error information
158 *
159 * Handle an internal error
160 */
161 static void LIBXML_ATTR_FORMAT(2,0)
xmlErrInternal(xmlParserCtxtPtr ctxt,const char * msg,const xmlChar * str)162 xmlErrInternal(xmlParserCtxtPtr ctxt, const char *msg, const xmlChar * str)
163 {
164 if ((ctxt != NULL) && (ctxt->disableSAX != 0) &&
165 (ctxt->instate == XML_PARSER_EOF))
166 return;
167 if (ctxt != NULL)
168 ctxt->errNo = XML_ERR_INTERNAL_ERROR;
169 __xmlRaiseError(NULL, NULL, NULL,
170 ctxt, NULL, XML_FROM_PARSER, XML_ERR_INTERNAL_ERROR,
171 XML_ERR_FATAL, NULL, 0, (const char *) str, NULL, NULL,
172 0, 0, msg, str);
173 if (ctxt != NULL) {
174 ctxt->wellFormed = 0;
175 if (ctxt->recovery == 0)
176 ctxt->disableSAX = 1;
177 }
178 }
179
180 /**
181 * xmlFatalErr:
182 * @ctxt: an XML parser context
183 * @error: the error number
184 * @info: extra information string
185 *
186 * Handle a fatal parser error, i.e. violating Well-Formedness constraints
187 */
188 void
xmlFatalErr(xmlParserCtxtPtr ctxt,xmlParserErrors error,const char * info)189 xmlFatalErr(xmlParserCtxtPtr ctxt, xmlParserErrors error, const char *info)
190 {
191 const char *errmsg;
192
193 if ((ctxt != NULL) && (ctxt->disableSAX != 0) &&
194 (ctxt->instate == XML_PARSER_EOF))
195 return;
196 switch (error) {
197 case XML_ERR_INVALID_HEX_CHARREF:
198 errmsg = "CharRef: invalid hexadecimal value";
199 break;
200 case XML_ERR_INVALID_DEC_CHARREF:
201 errmsg = "CharRef: invalid decimal value";
202 break;
203 case XML_ERR_INVALID_CHARREF:
204 errmsg = "CharRef: invalid value";
205 break;
206 case XML_ERR_INTERNAL_ERROR:
207 errmsg = "internal error";
208 break;
209 case XML_ERR_PEREF_AT_EOF:
210 errmsg = "PEReference at end of document";
211 break;
212 case XML_ERR_PEREF_IN_PROLOG:
213 errmsg = "PEReference in prolog";
214 break;
215 case XML_ERR_PEREF_IN_EPILOG:
216 errmsg = "PEReference in epilog";
217 break;
218 case XML_ERR_PEREF_NO_NAME:
219 errmsg = "PEReference: no name";
220 break;
221 case XML_ERR_PEREF_SEMICOL_MISSING:
222 errmsg = "PEReference: expecting ';'";
223 break;
224 case XML_ERR_ENTITY_LOOP:
225 errmsg = "Detected an entity reference loop";
226 break;
227 case XML_ERR_ENTITY_NOT_STARTED:
228 errmsg = "EntityValue: \" or ' expected";
229 break;
230 case XML_ERR_ENTITY_PE_INTERNAL:
231 errmsg = "PEReferences forbidden in internal subset";
232 break;
233 case XML_ERR_ENTITY_NOT_FINISHED:
234 errmsg = "EntityValue: \" or ' expected";
235 break;
236 case XML_ERR_ATTRIBUTE_NOT_STARTED:
237 errmsg = "AttValue: \" or ' expected";
238 break;
239 case XML_ERR_LT_IN_ATTRIBUTE:
240 errmsg = "Unescaped '<' not allowed in attributes values";
241 break;
242 case XML_ERR_LITERAL_NOT_STARTED:
243 errmsg = "SystemLiteral \" or ' expected";
244 break;
245 case XML_ERR_LITERAL_NOT_FINISHED:
246 errmsg = "Unfinished System or Public ID \" or ' expected";
247 break;
248 case XML_ERR_MISPLACED_CDATA_END:
249 errmsg = "Sequence ']]>' not allowed in content";
250 break;
251 case XML_ERR_URI_REQUIRED:
252 errmsg = "SYSTEM or PUBLIC, the URI is missing";
253 break;
254 case XML_ERR_PUBID_REQUIRED:
255 errmsg = "PUBLIC, the Public Identifier is missing";
256 break;
257 case XML_ERR_HYPHEN_IN_COMMENT:
258 errmsg = "Comment must not contain '--' (double-hyphen)";
259 break;
260 case XML_ERR_PI_NOT_STARTED:
261 errmsg = "xmlParsePI : no target name";
262 break;
263 case XML_ERR_RESERVED_XML_NAME:
264 errmsg = "Invalid PI name";
265 break;
266 case XML_ERR_NOTATION_NOT_STARTED:
267 errmsg = "NOTATION: Name expected here";
268 break;
269 case XML_ERR_NOTATION_NOT_FINISHED:
270 errmsg = "'>' required to close NOTATION declaration";
271 break;
272 case XML_ERR_VALUE_REQUIRED:
273 errmsg = "Entity value required";
274 break;
275 case XML_ERR_URI_FRAGMENT:
276 errmsg = "Fragment not allowed";
277 break;
278 case XML_ERR_ATTLIST_NOT_STARTED:
279 errmsg = "'(' required to start ATTLIST enumeration";
280 break;
281 case XML_ERR_NMTOKEN_REQUIRED:
282 errmsg = "NmToken expected in ATTLIST enumeration";
283 break;
284 case XML_ERR_ATTLIST_NOT_FINISHED:
285 errmsg = "')' required to finish ATTLIST enumeration";
286 break;
287 case XML_ERR_MIXED_NOT_STARTED:
288 errmsg = "MixedContentDecl : '|' or ')*' expected";
289 break;
290 case XML_ERR_PCDATA_REQUIRED:
291 errmsg = "MixedContentDecl : '#PCDATA' expected";
292 break;
293 case XML_ERR_ELEMCONTENT_NOT_STARTED:
294 errmsg = "ContentDecl : Name or '(' expected";
295 break;
296 case XML_ERR_ELEMCONTENT_NOT_FINISHED:
297 errmsg = "ContentDecl : ',' '|' or ')' expected";
298 break;
299 case XML_ERR_PEREF_IN_INT_SUBSET:
300 errmsg =
301 "PEReference: forbidden within markup decl in internal subset";
302 break;
303 case XML_ERR_GT_REQUIRED:
304 errmsg = "expected '>'";
305 break;
306 case XML_ERR_CONDSEC_INVALID:
307 errmsg = "XML conditional section '[' expected";
308 break;
309 case XML_ERR_EXT_SUBSET_NOT_FINISHED:
310 errmsg = "Content error in the external subset";
311 break;
312 case XML_ERR_CONDSEC_INVALID_KEYWORD:
313 errmsg =
314 "conditional section INCLUDE or IGNORE keyword expected";
315 break;
316 case XML_ERR_CONDSEC_NOT_FINISHED:
317 errmsg = "XML conditional section not closed";
318 break;
319 case XML_ERR_XMLDECL_NOT_STARTED:
320 errmsg = "Text declaration '<?xml' required";
321 break;
322 case XML_ERR_XMLDECL_NOT_FINISHED:
323 errmsg = "parsing XML declaration: '?>' expected";
324 break;
325 case XML_ERR_EXT_ENTITY_STANDALONE:
326 errmsg = "external parsed entities cannot be standalone";
327 break;
328 case XML_ERR_ENTITYREF_SEMICOL_MISSING:
329 errmsg = "EntityRef: expecting ';'";
330 break;
331 case XML_ERR_DOCTYPE_NOT_FINISHED:
332 errmsg = "DOCTYPE improperly terminated";
333 break;
334 case XML_ERR_LTSLASH_REQUIRED:
335 errmsg = "EndTag: '</' not found";
336 break;
337 case XML_ERR_EQUAL_REQUIRED:
338 errmsg = "expected '='";
339 break;
340 case XML_ERR_STRING_NOT_CLOSED:
341 errmsg = "String not closed expecting \" or '";
342 break;
343 case XML_ERR_STRING_NOT_STARTED:
344 errmsg = "String not started expecting ' or \"";
345 break;
346 case XML_ERR_ENCODING_NAME:
347 errmsg = "Invalid XML encoding name";
348 break;
349 case XML_ERR_STANDALONE_VALUE:
350 errmsg = "standalone accepts only 'yes' or 'no'";
351 break;
352 case XML_ERR_DOCUMENT_EMPTY:
353 errmsg = "Document is empty";
354 break;
355 case XML_ERR_DOCUMENT_END:
356 errmsg = "Extra content at the end of the document";
357 break;
358 case XML_ERR_NOT_WELL_BALANCED:
359 errmsg = "chunk is not well balanced";
360 break;
361 case XML_ERR_EXTRA_CONTENT:
362 errmsg = "extra content at the end of well balanced chunk";
363 break;
364 case XML_ERR_VERSION_MISSING:
365 errmsg = "Malformed declaration expecting version";
366 break;
367 case XML_ERR_NAME_TOO_LONG:
368 errmsg = "Name too long";
369 break;
370 case XML_ERR_INVALID_ENCODING:
371 errmsg = "Invalid bytes in character encoding";
372 break;
373 case XML_IO_UNKNOWN:
374 errmsg = "I/O error";
375 break;
376 #if 0
377 case:
378 errmsg = "";
379 break;
380 #endif
381 default:
382 errmsg = "Unregistered error message";
383 }
384 if (ctxt != NULL)
385 ctxt->errNo = error;
386 if (info == NULL) {
387 __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER, error,
388 XML_ERR_FATAL, NULL, 0, info, NULL, NULL, 0, 0, "%s\n",
389 errmsg);
390 } else {
391 __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER, error,
392 XML_ERR_FATAL, NULL, 0, info, NULL, NULL, 0, 0, "%s: %s\n",
393 errmsg, info);
394 }
395 if (ctxt != NULL) {
396 ctxt->wellFormed = 0;
397 if (ctxt->recovery == 0)
398 ctxt->disableSAX = 1;
399 }
400 }
401
402 /**
403 * xmlErrEncodingInt:
404 * @ctxt: an XML parser context
405 * @error: the error number
406 * @msg: the error message
407 * @val: an integer value
408 *
409 * n encoding error
410 */
411 static void LIBXML_ATTR_FORMAT(3,0)
xmlErrEncodingInt(xmlParserCtxtPtr ctxt,xmlParserErrors error,const char * msg,int val)412 xmlErrEncodingInt(xmlParserCtxtPtr ctxt, xmlParserErrors error,
413 const char *msg, int val)
414 {
415 if ((ctxt != NULL) && (ctxt->disableSAX != 0) &&
416 (ctxt->instate == XML_PARSER_EOF))
417 return;
418 if (ctxt != NULL)
419 ctxt->errNo = error;
420 __xmlRaiseError(NULL, NULL, NULL,
421 ctxt, NULL, XML_FROM_PARSER, error, XML_ERR_FATAL,
422 NULL, 0, NULL, NULL, NULL, val, 0, msg, val);
423 if (ctxt != NULL) {
424 ctxt->wellFormed = 0;
425 if (ctxt->recovery == 0)
426 ctxt->disableSAX = 1;
427 }
428 }
429
430 /**
431 * xmlIsLetter:
432 * @c: an unicode character (int)
433 *
434 * Check whether the character is allowed by the production
435 * [84] Letter ::= BaseChar | Ideographic
436 *
437 * Returns 0 if not, non-zero otherwise
438 */
439 int
xmlIsLetter(int c)440 xmlIsLetter(int c) {
441 return(IS_BASECHAR(c) || IS_IDEOGRAPHIC(c));
442 }
443
444 /************************************************************************
445 * *
446 * Input handling functions for progressive parsing *
447 * *
448 ************************************************************************/
449
450 /* we need to keep enough input to show errors in context */
451 #define LINE_LEN 80
452
453 /**
454 * xmlHaltParser:
455 * @ctxt: an XML parser context
456 *
457 * Blocks further parser processing don't override error
458 * for internal use
459 */
460 void
xmlHaltParser(xmlParserCtxtPtr ctxt)461 xmlHaltParser(xmlParserCtxtPtr ctxt) {
462 if (ctxt == NULL)
463 return;
464 ctxt->instate = XML_PARSER_EOF;
465 ctxt->disableSAX = 1;
466 while (ctxt->inputNr > 1)
467 xmlFreeInputStream(inputPop(ctxt));
468 if (ctxt->input != NULL) {
469 /*
470 * in case there was a specific allocation deallocate before
471 * overriding base
472 */
473 if (ctxt->input->free != NULL) {
474 ctxt->input->free((xmlChar *) ctxt->input->base);
475 ctxt->input->free = NULL;
476 }
477 if (ctxt->input->buf != NULL) {
478 xmlFreeParserInputBuffer(ctxt->input->buf);
479 ctxt->input->buf = NULL;
480 }
481 ctxt->input->cur = BAD_CAST"";
482 ctxt->input->length = 0;
483 ctxt->input->base = ctxt->input->cur;
484 ctxt->input->end = ctxt->input->cur;
485 }
486 }
487
488 /**
489 * xmlParserInputRead:
490 * @in: an XML parser input
491 * @len: an indicative size for the lookahead
492 *
493 * DEPRECATED: This function was internal and is deprecated.
494 *
495 * Returns -1 as this is an error to use it.
496 */
497 int
xmlParserInputRead(xmlParserInputPtr in ATTRIBUTE_UNUSED,int len ATTRIBUTE_UNUSED)498 xmlParserInputRead(xmlParserInputPtr in ATTRIBUTE_UNUSED, int len ATTRIBUTE_UNUSED) {
499 return(-1);
500 }
501
502 /**
503 * xmlParserGrow:
504 * @ctxt: an XML parser context
505 *
506 * Grow the input buffer.
507 *
508 * Returns the number of bytes read or -1 in case of error.
509 */
510 int
xmlParserGrow(xmlParserCtxtPtr ctxt)511 xmlParserGrow(xmlParserCtxtPtr ctxt) {
512 xmlParserInputPtr in = ctxt->input;
513 xmlParserInputBufferPtr buf = in->buf;
514 ptrdiff_t curEnd = in->end - in->cur;
515 ptrdiff_t curBase = in->cur - in->base;
516 int ret;
517
518 if (buf == NULL)
519 return(0);
520 /* Don't grow push parser buffer. */
521 if ((ctxt->progressive) && (ctxt->inputNr <= 1))
522 return(0);
523 if (buf->error != 0)
524 return(-1);
525
526 if (((curEnd > XML_MAX_LOOKUP_LIMIT) ||
527 (curBase > XML_MAX_LOOKUP_LIMIT)) &&
528 ((ctxt->options & XML_PARSE_HUGE) == 0)) {
529 xmlErrMemory(ctxt, "Huge input lookup");
530 xmlHaltParser(ctxt);
531 return(-1);
532 }
533
534 if (curEnd >= INPUT_CHUNK)
535 return(0);
536
537 ret = xmlParserInputBufferGrow(buf, INPUT_CHUNK);
538 xmlBufUpdateInput(buf->buffer, in, curBase);
539
540 if (ret < 0) {
541 xmlFatalErr(ctxt, buf->error, NULL);
542 /* Buffer contents may be lost in case of memory errors. */
543 if (buf->error == XML_ERR_NO_MEMORY)
544 xmlHaltParser(ctxt);
545 }
546
547 return(ret);
548 }
549
550 /**
551 * xmlParserInputGrow:
552 * @in: an XML parser input
553 * @len: an indicative size for the lookahead
554 *
555 * DEPRECATED: Don't use.
556 *
557 * This function increase the input for the parser. It tries to
558 * preserve pointers to the input buffer, and keep already read data
559 *
560 * Returns the amount of char read, or -1 in case of error, 0 indicate the
561 * end of this entity
562 */
563 int
xmlParserInputGrow(xmlParserInputPtr in,int len)564 xmlParserInputGrow(xmlParserInputPtr in, int len) {
565 int ret;
566 size_t indx;
567
568 if ((in == NULL) || (len < 0)) return(-1);
569 if (in->buf == NULL) return(-1);
570 if (in->base == NULL) return(-1);
571 if (in->cur == NULL) return(-1);
572 if (in->buf->buffer == NULL) return(-1);
573
574 indx = in->cur - in->base;
575 if (xmlBufUse(in->buf->buffer) > (unsigned int) indx + INPUT_CHUNK) {
576 return(0);
577 }
578 ret = xmlParserInputBufferGrow(in->buf, len);
579
580 in->base = xmlBufContent(in->buf->buffer);
581 if (in->base == NULL) {
582 in->base = BAD_CAST "";
583 in->cur = in->base;
584 in->end = in->base;
585 return(-1);
586 }
587 in->cur = in->base + indx;
588 in->end = xmlBufEnd(in->buf->buffer);
589
590 return(ret);
591 }
592
593 /**
594 * xmlParserShrink:
595 * @ctxt: an XML parser context
596 *
597 * Shrink the input buffer.
598 */
599 void
xmlParserShrink(xmlParserCtxtPtr ctxt)600 xmlParserShrink(xmlParserCtxtPtr ctxt) {
601 xmlParserInputPtr in = ctxt->input;
602 xmlParserInputBufferPtr buf = in->buf;
603 size_t used;
604
605 if (buf == NULL)
606 return;
607
608 used = in->cur - in->base;
609 /*
610 * Do not shrink on large buffers whose only a tiny fraction
611 * was consumed
612 */
613 if (used > INPUT_CHUNK) {
614 size_t res = xmlBufShrink(buf->buffer, used - LINE_LEN);
615
616 if (res > 0) {
617 used -= res;
618 if ((res > ULONG_MAX) ||
619 (in->consumed > ULONG_MAX - (unsigned long)res))
620 in->consumed = ULONG_MAX;
621 else
622 in->consumed += res;
623 }
624 }
625
626 xmlBufUpdateInput(buf->buffer, in, used);
627 }
628
629 /**
630 * xmlParserInputShrink:
631 * @in: an XML parser input
632 *
633 * DEPRECATED: Don't use.
634 *
635 * This function removes used input for the parser.
636 */
637 void
xmlParserInputShrink(xmlParserInputPtr in)638 xmlParserInputShrink(xmlParserInputPtr in) {
639 size_t used;
640 size_t ret;
641
642 if (in == NULL) return;
643 if (in->buf == NULL) return;
644 if (in->base == NULL) return;
645 if (in->cur == NULL) return;
646 if (in->buf->buffer == NULL) return;
647
648 used = in->cur - in->base;
649 /*
650 * Do not shrink on large buffers whose only a tiny fraction
651 * was consumed
652 */
653 if (used > INPUT_CHUNK) {
654 ret = xmlBufShrink(in->buf->buffer, used - LINE_LEN);
655 if (ret > 0) {
656 used -= ret;
657 if ((ret > ULONG_MAX) ||
658 (in->consumed > ULONG_MAX - (unsigned long)ret))
659 in->consumed = ULONG_MAX;
660 else
661 in->consumed += ret;
662 }
663 }
664
665 if (xmlBufUse(in->buf->buffer) <= INPUT_CHUNK) {
666 xmlParserInputBufferRead(in->buf, 2 * INPUT_CHUNK);
667 }
668
669 in->base = xmlBufContent(in->buf->buffer);
670 if (in->base == NULL) {
671 /* TODO: raise error */
672 in->base = BAD_CAST "";
673 in->cur = in->base;
674 in->end = in->base;
675 return;
676 }
677 in->cur = in->base + used;
678 in->end = xmlBufEnd(in->buf->buffer);
679 }
680
681 /************************************************************************
682 * *
683 * UTF8 character input and related functions *
684 * *
685 ************************************************************************/
686
687 /**
688 * xmlNextChar:
689 * @ctxt: the XML parser context
690 *
691 * DEPRECATED: Internal function, do not use.
692 *
693 * Skip to the next char input char.
694 */
695
696 void
xmlNextChar(xmlParserCtxtPtr ctxt)697 xmlNextChar(xmlParserCtxtPtr ctxt)
698 {
699 const unsigned char *cur;
700 size_t avail;
701 int c;
702
703 if ((ctxt == NULL) || (ctxt->instate == XML_PARSER_EOF) ||
704 (ctxt->input == NULL))
705 return;
706
707 avail = ctxt->input->end - ctxt->input->cur;
708
709 if (avail < INPUT_CHUNK) {
710 xmlParserGrow(ctxt);
711 if ((ctxt->instate == XML_PARSER_EOF) ||
712 (ctxt->input->cur >= ctxt->input->end))
713 return;
714 avail = ctxt->input->end - ctxt->input->cur;
715 }
716
717 cur = ctxt->input->cur;
718 c = *cur;
719
720 if (c < 0x80) {
721 if (c == '\n') {
722 ctxt->input->cur++;
723 ctxt->input->line++;
724 ctxt->input->col = 1;
725 } else if (c == '\r') {
726 /*
727 * 2.11 End-of-Line Handling
728 * the literal two-character sequence "#xD#xA" or a standalone
729 * literal #xD, an XML processor must pass to the application
730 * the single character #xA.
731 */
732 ctxt->input->cur += ((cur[1] == '\n') ? 2 : 1);
733 ctxt->input->line++;
734 ctxt->input->col = 1;
735 return;
736 } else {
737 ctxt->input->cur++;
738 ctxt->input->col++;
739 }
740 } else {
741 ctxt->input->col++;
742
743 if ((avail < 2) || (cur[1] & 0xc0) != 0x80)
744 goto encoding_error;
745
746 if (c < 0xe0) {
747 /* 2-byte code */
748 if (c < 0xc2)
749 goto encoding_error;
750 ctxt->input->cur += 2;
751 } else {
752 unsigned int val = (c << 8) | cur[1];
753
754 if ((avail < 3) || (cur[2] & 0xc0) != 0x80)
755 goto encoding_error;
756
757 if (c < 0xf0) {
758 /* 3-byte code */
759 if ((val < 0xe0a0) || ((val >= 0xeda0) && (val < 0xee00)))
760 goto encoding_error;
761 ctxt->input->cur += 3;
762 } else {
763 if ((avail < 4) || ((cur[3] & 0xc0) != 0x80))
764 goto encoding_error;
765
766 /* 4-byte code */
767 if ((val < 0xf090) || (val >= 0xf490))
768 goto encoding_error;
769 ctxt->input->cur += 4;
770 }
771 }
772 }
773
774 return;
775
776 encoding_error:
777 /* Only report the first error */
778 if ((ctxt->input->flags & XML_INPUT_ENCODING_ERROR) == 0) {
779 if ((ctxt == NULL) || (ctxt->input == NULL) ||
780 (ctxt->input->end - ctxt->input->cur < 4)) {
781 __xmlErrEncoding(ctxt, XML_ERR_INVALID_CHAR,
782 "Input is not proper UTF-8, indicate encoding !\n",
783 NULL, NULL);
784 } else {
785 char buffer[150];
786
787 snprintf(buffer, 149, "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n",
788 ctxt->input->cur[0], ctxt->input->cur[1],
789 ctxt->input->cur[2], ctxt->input->cur[3]);
790 __xmlErrEncoding(ctxt, XML_ERR_INVALID_CHAR,
791 "Input is not proper UTF-8, indicate encoding !\n%s",
792 BAD_CAST buffer, NULL);
793 }
794 ctxt->input->flags |= XML_INPUT_ENCODING_ERROR;
795 }
796 ctxt->input->cur++;
797 return;
798 }
799
800 /**
801 * xmlCurrentChar:
802 * @ctxt: the XML parser context
803 * @len: pointer to the length of the char read
804 *
805 * DEPRECATED: Internal function, do not use.
806 *
807 * The current char value, if using UTF-8 this may actually span multiple
808 * bytes in the input buffer. Implement the end of line normalization:
809 * 2.11 End-of-Line Handling
810 * Wherever an external parsed entity or the literal entity value
811 * of an internal parsed entity contains either the literal two-character
812 * sequence "#xD#xA" or a standalone literal #xD, an XML processor
813 * must pass to the application the single character #xA.
814 * This behavior can conveniently be produced by normalizing all
815 * line breaks to #xA on input, before parsing.)
816 *
817 * Returns the current char value and its length
818 */
819
820 int
xmlCurrentChar(xmlParserCtxtPtr ctxt,int * len)821 xmlCurrentChar(xmlParserCtxtPtr ctxt, int *len) {
822 const unsigned char *cur;
823 size_t avail;
824 int c;
825
826 if ((ctxt == NULL) || (len == NULL) || (ctxt->input == NULL)) return(0);
827 if (ctxt->instate == XML_PARSER_EOF)
828 return(0);
829
830 avail = ctxt->input->end - ctxt->input->cur;
831
832 if (avail < INPUT_CHUNK) {
833 xmlParserGrow(ctxt);
834 if (ctxt->instate == XML_PARSER_EOF)
835 return(0);
836 avail = ctxt->input->end - ctxt->input->cur;
837 }
838
839 cur = ctxt->input->cur;
840 c = *cur;
841
842 if (c < 0x80) {
843 /* 1-byte code */
844 if (c < 0x20) {
845 /*
846 * 2.11 End-of-Line Handling
847 * the literal two-character sequence "#xD#xA" or a standalone
848 * literal #xD, an XML processor must pass to the application
849 * the single character #xA.
850 */
851 if (c == '\r') {
852 *len = ((cur[1] == '\n') ? 2 : 1);
853 c = '\n';
854 } else if (c == 0) {
855 if (ctxt->input->cur >= ctxt->input->end) {
856 *len = 0;
857 } else {
858 *len = 1;
859 /*
860 * TODO: Null bytes should be handled by callers,
861 * but this can be tricky.
862 */
863 xmlErrEncodingInt(ctxt, XML_ERR_INVALID_CHAR,
864 "Char 0x0 out of allowed range\n", c);
865 }
866 } else {
867 *len = 1;
868 }
869 } else {
870 *len = 1;
871 }
872
873 return(c);
874 } else {
875 int val;
876
877 if (avail < 2)
878 goto incomplete_sequence;
879 if ((cur[1] & 0xc0) != 0x80)
880 goto encoding_error;
881
882 if (c < 0xe0) {
883 /* 2-byte code */
884 if (c < 0xc2)
885 goto encoding_error;
886 val = (c & 0x1f) << 6;
887 val |= cur[1] & 0x3f;
888 *len = 2;
889 } else {
890 if (avail < 3)
891 goto incomplete_sequence;
892 if ((cur[2] & 0xc0) != 0x80)
893 goto encoding_error;
894
895 if (c < 0xf0) {
896 /* 3-byte code */
897 val = (c & 0xf) << 12;
898 val |= (cur[1] & 0x3f) << 6;
899 val |= cur[2] & 0x3f;
900 if ((val < 0x800) || ((val >= 0xd800) && (val < 0xe000)))
901 goto encoding_error;
902 *len = 3;
903 } else {
904 if (avail < 4)
905 goto incomplete_sequence;
906 if ((cur[3] & 0xc0) != 0x80)
907 goto encoding_error;
908
909 /* 4-byte code */
910 val = (c & 0x0f) << 18;
911 val |= (cur[1] & 0x3f) << 12;
912 val |= (cur[2] & 0x3f) << 6;
913 val |= cur[3] & 0x3f;
914 if ((val < 0x10000) || (val >= 0x110000))
915 goto encoding_error;
916 *len = 4;
917 }
918 }
919
920 return(val);
921 }
922
923 encoding_error:
924 /* Only report the first error */
925 if ((ctxt->input->flags & XML_INPUT_ENCODING_ERROR) == 0) {
926 if (ctxt->input->end - ctxt->input->cur < 4) {
927 __xmlErrEncoding(ctxt, XML_ERR_INVALID_CHAR,
928 "Input is not proper UTF-8, indicate encoding !\n",
929 NULL, NULL);
930 } else {
931 char buffer[150];
932
933 snprintf(&buffer[0], 149, "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n",
934 ctxt->input->cur[0], ctxt->input->cur[1],
935 ctxt->input->cur[2], ctxt->input->cur[3]);
936 __xmlErrEncoding(ctxt, XML_ERR_INVALID_CHAR,
937 "Input is not proper UTF-8, indicate encoding !\n%s",
938 BAD_CAST buffer, NULL);
939 }
940 ctxt->input->flags |= XML_INPUT_ENCODING_ERROR;
941 }
942 *len = 1;
943 return(0xFFFD); /* U+FFFD Replacement Character */
944
945 incomplete_sequence:
946 /*
947 * An encoding problem may arise from a truncated input buffer
948 * splitting a character in the middle. In that case do not raise
949 * an error but return 0. This should only happen when push parsing
950 * char data.
951 */
952 *len = 0;
953 return(0);
954 }
955
956 /**
957 * xmlStringCurrentChar:
958 * @ctxt: the XML parser context
959 * @cur: pointer to the beginning of the char
960 * @len: pointer to the length of the char read
961 *
962 * DEPRECATED: Internal function, do not use.
963 *
964 * The current char value, if using UTF-8 this may actually span multiple
965 * bytes in the input buffer.
966 *
967 * Returns the current char value and its length
968 */
969
970 int
xmlStringCurrentChar(xmlParserCtxtPtr ctxt ATTRIBUTE_UNUSED,const xmlChar * cur,int * len)971 xmlStringCurrentChar(xmlParserCtxtPtr ctxt ATTRIBUTE_UNUSED,
972 const xmlChar *cur, int *len) {
973 int c;
974
975 if ((cur == NULL) || (len == NULL))
976 return(0);
977
978 /* cur is zero-terminated, so we can lie about its length. */
979 *len = 4;
980 c = xmlGetUTF8Char(cur, len);
981
982 return((c < 0) ? 0 : c);
983 }
984
985 /**
986 * xmlCopyCharMultiByte:
987 * @out: pointer to an array of xmlChar
988 * @val: the char value
989 *
990 * append the char value in the array
991 *
992 * Returns the number of xmlChar written
993 */
994 int
xmlCopyCharMultiByte(xmlChar * out,int val)995 xmlCopyCharMultiByte(xmlChar *out, int val) {
996 if ((out == NULL) || (val < 0)) return(0);
997 /*
998 * We are supposed to handle UTF8, check it's valid
999 * From rfc2044: encoding of the Unicode values on UTF-8:
1000 *
1001 * UCS-4 range (hex.) UTF-8 octet sequence (binary)
1002 * 0000 0000-0000 007F 0xxxxxxx
1003 * 0000 0080-0000 07FF 110xxxxx 10xxxxxx
1004 * 0000 0800-0000 FFFF 1110xxxx 10xxxxxx 10xxxxxx
1005 */
1006 if (val >= 0x80) {
1007 xmlChar *savedout = out;
1008 int bits;
1009 if (val < 0x800) { *out++= (val >> 6) | 0xC0; bits= 0; }
1010 else if (val < 0x10000) { *out++= (val >> 12) | 0xE0; bits= 6;}
1011 else if (val < 0x110000) { *out++= (val >> 18) | 0xF0; bits= 12; }
1012 else {
1013 xmlErrEncodingInt(NULL, XML_ERR_INVALID_CHAR,
1014 "Internal error, xmlCopyCharMultiByte 0x%X out of bound\n",
1015 val);
1016 return(0);
1017 }
1018 for ( ; bits >= 0; bits-= 6)
1019 *out++= ((val >> bits) & 0x3F) | 0x80 ;
1020 return (out - savedout);
1021 }
1022 *out = val;
1023 return 1;
1024 }
1025
1026 /**
1027 * xmlCopyChar:
1028 * @len: Ignored, compatibility
1029 * @out: pointer to an array of xmlChar
1030 * @val: the char value
1031 *
1032 * append the char value in the array
1033 *
1034 * Returns the number of xmlChar written
1035 */
1036
1037 int
xmlCopyChar(int len ATTRIBUTE_UNUSED,xmlChar * out,int val)1038 xmlCopyChar(int len ATTRIBUTE_UNUSED, xmlChar *out, int val) {
1039 if ((out == NULL) || (val < 0)) return(0);
1040 /* the len parameter is ignored */
1041 if (val >= 0x80) {
1042 return(xmlCopyCharMultiByte (out, val));
1043 }
1044 *out = val;
1045 return 1;
1046 }
1047
1048 /************************************************************************
1049 * *
1050 * Commodity functions to switch encodings *
1051 * *
1052 ************************************************************************/
1053
1054 static xmlCharEncodingHandlerPtr
xmlDetectEBCDIC(xmlParserInputPtr input)1055 xmlDetectEBCDIC(xmlParserInputPtr input) {
1056 xmlChar out[200];
1057 xmlCharEncodingHandlerPtr handler;
1058 int inlen, outlen, res, i;
1059
1060 /*
1061 * To detect the EBCDIC code page, we convert the first 200 bytes
1062 * to EBCDIC-US and try to find the encoding declaration.
1063 */
1064 handler = xmlGetCharEncodingHandler(XML_CHAR_ENCODING_EBCDIC);
1065 if (handler == NULL)
1066 return(NULL);
1067 outlen = sizeof(out) - 1;
1068 inlen = input->end - input->cur;
1069 res = xmlEncInputChunk(handler, out, &outlen, input->cur, &inlen);
1070 if (res < 0)
1071 return(handler);
1072 out[outlen] = 0;
1073
1074 for (i = 0; i < outlen; i++) {
1075 if (out[i] == '>')
1076 break;
1077 if ((out[i] == 'e') &&
1078 (xmlStrncmp(out + i, BAD_CAST "encoding", 8) == 0)) {
1079 int start, cur, quote;
1080
1081 i += 8;
1082 while (IS_BLANK_CH(out[i]))
1083 i += 1;
1084 if (out[i++] != '=')
1085 break;
1086 while (IS_BLANK_CH(out[i]))
1087 i += 1;
1088 quote = out[i++];
1089 if ((quote != '\'') && (quote != '"'))
1090 break;
1091 start = i;
1092 cur = out[i];
1093 while (((cur >= 'a') && (cur <= 'z')) ||
1094 ((cur >= 'A') && (cur <= 'Z')) ||
1095 ((cur >= '0') && (cur <= '9')) ||
1096 (cur == '.') || (cur == '_') ||
1097 (cur == '-'))
1098 cur = out[++i];
1099 if (cur != quote)
1100 break;
1101 out[i] = 0;
1102 xmlCharEncCloseFunc(handler);
1103 return(xmlFindCharEncodingHandler((char *) out + start));
1104 }
1105 }
1106
1107 /*
1108 * ICU handlers are stateful, so we have to recreate them.
1109 */
1110 xmlCharEncCloseFunc(handler);
1111 return(xmlGetCharEncodingHandler(XML_CHAR_ENCODING_EBCDIC));
1112 }
1113
1114 /**
1115 * xmlSwitchEncoding:
1116 * @ctxt: the parser context
1117 * @enc: the encoding value (number)
1118 *
1119 * Use encoding specified by enum to decode input data.
1120 *
1121 * This function can be used to enforce the encoding of chunks passed
1122 * to xmlParseChunk.
1123 *
1124 * Returns 0 in case of success, -1 otherwise
1125 */
1126 int
xmlSwitchEncoding(xmlParserCtxtPtr ctxt,xmlCharEncoding enc)1127 xmlSwitchEncoding(xmlParserCtxtPtr ctxt, xmlCharEncoding enc)
1128 {
1129 xmlCharEncodingHandlerPtr handler = NULL;
1130 int check = 1;
1131 int ret;
1132
1133 if ((ctxt == NULL) || (ctxt->input == NULL))
1134 return(-1);
1135
1136 switch (enc) {
1137 case XML_CHAR_ENCODING_NONE:
1138 case XML_CHAR_ENCODING_UTF8:
1139 case XML_CHAR_ENCODING_ASCII:
1140 check = 0;
1141 break;
1142 case XML_CHAR_ENCODING_EBCDIC:
1143 handler = xmlDetectEBCDIC(ctxt->input);
1144 break;
1145 default:
1146 handler = xmlGetCharEncodingHandler(enc);
1147 break;
1148 }
1149
1150 if ((check) && (handler == NULL)) {
1151 const char *name = xmlGetCharEncodingName(enc);
1152
1153 __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
1154 "encoding not supported: %s\n",
1155 BAD_CAST (name ? name : "<null>"), NULL);
1156 /*
1157 * TODO: We could recover from errors in external entities
1158 * if we didn't stop the parser. But most callers of this
1159 * function don't check the return value.
1160 */
1161 xmlStopParser(ctxt);
1162 return(-1);
1163 }
1164
1165 ret = xmlSwitchInputEncoding(ctxt, ctxt->input, handler);
1166
1167 if ((ret >= 0) && (enc == XML_CHAR_ENCODING_NONE)) {
1168 ctxt->input->flags &= ~XML_INPUT_HAS_ENCODING;
1169 }
1170
1171 return(ret);
1172 }
1173
1174 /**
1175 * xmlSwitchInputEncoding:
1176 * @ctxt: the parser context
1177 * @input: the input stream
1178 * @handler: the encoding handler
1179 *
1180 * DEPRECATED: Internal function, don't use.
1181 *
1182 * Use encoding handler to decode input data.
1183 *
1184 * Returns 0 in case of success, -1 otherwise
1185 */
1186 int
xmlSwitchInputEncoding(xmlParserCtxtPtr ctxt,xmlParserInputPtr input,xmlCharEncodingHandlerPtr handler)1187 xmlSwitchInputEncoding(xmlParserCtxtPtr ctxt, xmlParserInputPtr input,
1188 xmlCharEncodingHandlerPtr handler)
1189 {
1190 int nbchars;
1191 xmlParserInputBufferPtr in;
1192
1193 if ((input == NULL) || (input->buf == NULL)) {
1194 xmlCharEncCloseFunc(handler);
1195 return (-1);
1196 }
1197 in = input->buf;
1198
1199 input->flags |= XML_INPUT_HAS_ENCODING;
1200
1201 /*
1202 * UTF-8 requires no encoding handler.
1203 */
1204 if ((handler != NULL) &&
1205 (xmlStrcasecmp(BAD_CAST handler->name, BAD_CAST "UTF-8") == 0)) {
1206 xmlCharEncCloseFunc(handler);
1207 handler = NULL;
1208 }
1209
1210 if (in->encoder == handler)
1211 return (0);
1212
1213 if (in->encoder != NULL) {
1214 /*
1215 * Switching encodings during parsing is a really bad idea,
1216 * but Chromium can switch between ISO-8859-1 and UTF-16 before
1217 * separate calls to xmlParseChunk.
1218 *
1219 * TODO: We should check whether the "raw" input buffer is empty and
1220 * convert the old content using the old encoder.
1221 */
1222
1223 xmlCharEncCloseFunc(in->encoder);
1224 in->encoder = handler;
1225 return (0);
1226 }
1227
1228 in->encoder = handler;
1229
1230 /*
1231 * Is there already some content down the pipe to convert ?
1232 */
1233 if (xmlBufIsEmpty(in->buffer) == 0) {
1234 size_t processed;
1235
1236 /*
1237 * Shrink the current input buffer.
1238 * Move it as the raw buffer and create a new input buffer
1239 */
1240 processed = input->cur - input->base;
1241 xmlBufShrink(in->buffer, processed);
1242 input->consumed += processed;
1243 in->raw = in->buffer;
1244 in->buffer = xmlBufCreate();
1245 in->rawconsumed = processed;
1246
1247 nbchars = xmlCharEncInput(in);
1248 xmlBufResetInput(in->buffer, input);
1249 if (nbchars < 0) {
1250 /* TODO: This could be an out of memory or an encoding error. */
1251 xmlErrInternal(ctxt,
1252 "switching encoding: encoder error\n",
1253 NULL);
1254 xmlHaltParser(ctxt);
1255 return (-1);
1256 }
1257 }
1258 return (0);
1259 }
1260
1261 /**
1262 * xmlSwitchToEncoding:
1263 * @ctxt: the parser context
1264 * @handler: the encoding handler
1265 *
1266 * Use encoding handler to decode input data.
1267 *
1268 * This function can be used to enforce the encoding of chunks passed
1269 * to xmlParseChunk.
1270 *
1271 * Returns 0 in case of success, -1 otherwise
1272 */
1273 int
xmlSwitchToEncoding(xmlParserCtxtPtr ctxt,xmlCharEncodingHandlerPtr handler)1274 xmlSwitchToEncoding(xmlParserCtxtPtr ctxt, xmlCharEncodingHandlerPtr handler)
1275 {
1276 if (ctxt == NULL)
1277 return(-1);
1278 return(xmlSwitchInputEncoding(ctxt, ctxt->input, handler));
1279 }
1280
1281 /**
1282 * xmlDetectEncoding:
1283 * @ctxt: the parser context
1284 *
1285 * Handle optional BOM, detect and switch to encoding.
1286 *
1287 * Assumes that there are at least four bytes in the input buffer.
1288 */
1289 void
xmlDetectEncoding(xmlParserCtxtPtr ctxt)1290 xmlDetectEncoding(xmlParserCtxtPtr ctxt) {
1291 const xmlChar *in;
1292 xmlCharEncoding enc;
1293 int bomSize;
1294 int autoFlag = 0;
1295
1296 if (xmlParserGrow(ctxt) < 0)
1297 return;
1298 in = ctxt->input->cur;
1299 if (ctxt->input->end - in < 4)
1300 return;
1301
1302 if (ctxt->input->flags & XML_INPUT_HAS_ENCODING) {
1303 /*
1304 * If the encoding was already set, only skip the BOM which was
1305 * possibly decoded to UTF-8.
1306 */
1307 if ((in[0] == 0xEF) && (in[1] == 0xBB) && (in[2] == 0xBF)) {
1308 ctxt->input->cur += 3;
1309 }
1310
1311 return;
1312 }
1313
1314 enc = XML_CHAR_ENCODING_NONE;
1315 bomSize = 0;
1316
1317 switch (in[0]) {
1318 case 0x00:
1319 if ((in[1] == 0x00) && (in[2] == 0x00) && (in[3] == 0x3C)) {
1320 enc = XML_CHAR_ENCODING_UCS4BE;
1321 autoFlag = XML_INPUT_AUTO_OTHER;
1322 } else if ((in[1] == 0x3C) && (in[2] == 0x00) && (in[3] == 0x3F)) {
1323 enc = XML_CHAR_ENCODING_UTF16BE;
1324 autoFlag = XML_INPUT_AUTO_UTF16BE;
1325 }
1326 break;
1327
1328 case 0x3C:
1329 if (in[1] == 0x00) {
1330 if ((in[2] == 0x00) && (in[3] == 0x00)) {
1331 enc = XML_CHAR_ENCODING_UCS4LE;
1332 autoFlag = XML_INPUT_AUTO_OTHER;
1333 } else if ((in[2] == 0x3F) && (in[3] == 0x00)) {
1334 enc = XML_CHAR_ENCODING_UTF16LE;
1335 autoFlag = XML_INPUT_AUTO_UTF16LE;
1336 }
1337 }
1338 break;
1339
1340 case 0x4C:
1341 if ((in[1] == 0x6F) && (in[2] == 0xA7) && (in[3] == 0x94)) {
1342 enc = XML_CHAR_ENCODING_EBCDIC;
1343 autoFlag = XML_INPUT_AUTO_OTHER;
1344 }
1345 break;
1346
1347 case 0xEF:
1348 if ((in[1] == 0xBB) && (in[2] == 0xBF)) {
1349 enc = XML_CHAR_ENCODING_UTF8;
1350 autoFlag = XML_INPUT_AUTO_UTF8;
1351 bomSize = 3;
1352 }
1353 break;
1354
1355 case 0xFE:
1356 if (in[1] == 0xFF) {
1357 enc = XML_CHAR_ENCODING_UTF16BE;
1358 autoFlag = XML_INPUT_AUTO_UTF16BE;
1359 bomSize = 2;
1360 }
1361 break;
1362
1363 case 0xFF:
1364 if (in[1] == 0xFE) {
1365 enc = XML_CHAR_ENCODING_UTF16LE;
1366 autoFlag = XML_INPUT_AUTO_UTF16LE;
1367 bomSize = 2;
1368 }
1369 break;
1370 }
1371
1372 if (bomSize > 0) {
1373 ctxt->input->cur += bomSize;
1374 }
1375
1376 if (enc != XML_CHAR_ENCODING_NONE) {
1377 ctxt->input->flags |= autoFlag;
1378 xmlSwitchEncoding(ctxt, enc);
1379 }
1380 }
1381
1382 /**
1383 * xmlSetDeclaredEncoding:
1384 * @ctxt: the parser context
1385 * @encoding: declared encoding
1386 *
1387 * Set the encoding from a declaration in the document.
1388 *
1389 * If no encoding was set yet, switch the encoding. Otherwise, only warn
1390 * about encoding mismatches.
1391 *
1392 * Takes ownership of 'encoding'.
1393 */
1394 void
xmlSetDeclaredEncoding(xmlParserCtxtPtr ctxt,xmlChar * encoding)1395 xmlSetDeclaredEncoding(xmlParserCtxtPtr ctxt, xmlChar *encoding) {
1396 if (ctxt->encoding != NULL)
1397 xmlFree((xmlChar *) ctxt->encoding);
1398 ctxt->encoding = encoding;
1399
1400 if (((ctxt->input->flags & XML_INPUT_HAS_ENCODING) == 0) &&
1401 ((ctxt->options & XML_PARSE_IGNORE_ENC) == 0)) {
1402 xmlCharEncodingHandlerPtr handler;
1403
1404 handler = xmlFindCharEncodingHandler((const char *) encoding);
1405 if (handler == NULL) {
1406 __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
1407 "Unsupported encoding: %s\n",
1408 encoding, NULL);
1409 return;
1410 }
1411
1412 xmlSwitchToEncoding(ctxt, handler);
1413 ctxt->input->flags |= XML_INPUT_USES_ENC_DECL;
1414 } else if (ctxt->input->flags & XML_INPUT_AUTO_ENCODING) {
1415 static const char *allowedUTF8[] = {
1416 "UTF-8", "UTF8", NULL
1417 };
1418 static const char *allowedUTF16LE[] = {
1419 "UTF-16", "UTF-16LE", "UTF16", NULL
1420 };
1421 static const char *allowedUTF16BE[] = {
1422 "UTF-16", "UTF-16BE", "UTF16", NULL
1423 };
1424 const char **allowed = NULL;
1425 const char *autoEnc = NULL;
1426
1427 switch (ctxt->input->flags & XML_INPUT_AUTO_ENCODING) {
1428 case XML_INPUT_AUTO_UTF8:
1429 allowed = allowedUTF8;
1430 autoEnc = "UTF-8";
1431 break;
1432 case XML_INPUT_AUTO_UTF16LE:
1433 allowed = allowedUTF16LE;
1434 autoEnc = "UTF-16LE";
1435 break;
1436 case XML_INPUT_AUTO_UTF16BE:
1437 allowed = allowedUTF16BE;
1438 autoEnc = "UTF-16BE";
1439 break;
1440 }
1441
1442 if (allowed != NULL) {
1443 const char **p;
1444 int match = 0;
1445
1446 for (p = allowed; *p != NULL; p++) {
1447 if (xmlStrcasecmp(encoding, BAD_CAST *p) == 0) {
1448 match = 1;
1449 break;
1450 }
1451 }
1452
1453 if (match == 0) {
1454 xmlWarningMsg(ctxt, XML_WAR_ENCODING_MISMATCH,
1455 "Encoding '%s' doesn't match "
1456 "auto-detected '%s'\n",
1457 encoding, BAD_CAST autoEnc);
1458 }
1459 }
1460 }
1461 }
1462
1463 /************************************************************************
1464 * *
1465 * Commodity functions to handle entities processing *
1466 * *
1467 ************************************************************************/
1468
1469 /**
1470 * xmlFreeInputStream:
1471 * @input: an xmlParserInputPtr
1472 *
1473 * Free up an input stream.
1474 */
1475 void
xmlFreeInputStream(xmlParserInputPtr input)1476 xmlFreeInputStream(xmlParserInputPtr input) {
1477 if (input == NULL) return;
1478
1479 if (input->filename != NULL) xmlFree((char *) input->filename);
1480 if (input->directory != NULL) xmlFree((char *) input->directory);
1481 if (input->version != NULL) xmlFree((char *) input->version);
1482 if ((input->free != NULL) && (input->base != NULL))
1483 input->free((xmlChar *) input->base);
1484 if (input->buf != NULL)
1485 xmlFreeParserInputBuffer(input->buf);
1486 xmlFree(input);
1487 }
1488
1489 /**
1490 * xmlNewInputStream:
1491 * @ctxt: an XML parser context
1492 *
1493 * Create a new input stream structure.
1494 *
1495 * Returns the new input stream or NULL
1496 */
1497 xmlParserInputPtr
xmlNewInputStream(xmlParserCtxtPtr ctxt)1498 xmlNewInputStream(xmlParserCtxtPtr ctxt) {
1499 xmlParserInputPtr input;
1500
1501 input = (xmlParserInputPtr) xmlMalloc(sizeof(xmlParserInput));
1502 if (input == NULL) {
1503 xmlErrMemory(ctxt, "couldn't allocate a new input stream\n");
1504 return(NULL);
1505 }
1506 memset(input, 0, sizeof(xmlParserInput));
1507 input->line = 1;
1508 input->col = 1;
1509
1510 /*
1511 * If the context is NULL the id cannot be initialized, but that
1512 * should not happen while parsing which is the situation where
1513 * the id is actually needed.
1514 */
1515 if (ctxt != NULL) {
1516 if (input->id >= INT_MAX) {
1517 xmlErrMemory(ctxt, "Input ID overflow\n");
1518 return(NULL);
1519 }
1520 input->id = ctxt->input_id++;
1521 }
1522
1523 return(input);
1524 }
1525
1526 /**
1527 * xmlNewIOInputStream:
1528 * @ctxt: an XML parser context
1529 * @input: an I/O Input
1530 * @enc: the charset encoding if known
1531 *
1532 * Create a new input stream structure encapsulating the @input into
1533 * a stream suitable for the parser.
1534 *
1535 * Returns the new input stream or NULL
1536 */
1537 xmlParserInputPtr
xmlNewIOInputStream(xmlParserCtxtPtr ctxt,xmlParserInputBufferPtr input,xmlCharEncoding enc)1538 xmlNewIOInputStream(xmlParserCtxtPtr ctxt, xmlParserInputBufferPtr input,
1539 xmlCharEncoding enc) {
1540 xmlParserInputPtr inputStream;
1541
1542 if (input == NULL) return(NULL);
1543 if (xmlParserDebugEntities)
1544 xmlGenericError(xmlGenericErrorContext, "new input from I/O\n");
1545 inputStream = xmlNewInputStream(ctxt);
1546 if (inputStream == NULL) {
1547 return(NULL);
1548 }
1549 inputStream->filename = NULL;
1550 inputStream->buf = input;
1551 xmlBufResetInput(inputStream->buf->buffer, inputStream);
1552
1553 if (enc != XML_CHAR_ENCODING_NONE) {
1554 xmlSwitchEncoding(ctxt, enc);
1555 }
1556
1557 return(inputStream);
1558 }
1559
1560 /**
1561 * xmlNewEntityInputStream:
1562 * @ctxt: an XML parser context
1563 * @entity: an Entity pointer
1564 *
1565 * DEPRECATED: Internal function, do not use.
1566 *
1567 * Create a new input stream based on an xmlEntityPtr
1568 *
1569 * Returns the new input stream or NULL
1570 */
1571 xmlParserInputPtr
xmlNewEntityInputStream(xmlParserCtxtPtr ctxt,xmlEntityPtr entity)1572 xmlNewEntityInputStream(xmlParserCtxtPtr ctxt, xmlEntityPtr entity) {
1573 xmlParserInputPtr input;
1574
1575 if (entity == NULL) {
1576 xmlErrInternal(ctxt, "xmlNewEntityInputStream entity = NULL\n",
1577 NULL);
1578 return(NULL);
1579 }
1580 if (xmlParserDebugEntities)
1581 xmlGenericError(xmlGenericErrorContext,
1582 "new input from entity: %s\n", entity->name);
1583 if (entity->content == NULL) {
1584 switch (entity->etype) {
1585 case XML_EXTERNAL_GENERAL_UNPARSED_ENTITY:
1586 xmlErrInternal(ctxt, "Cannot parse entity %s\n",
1587 entity->name);
1588 break;
1589 case XML_EXTERNAL_GENERAL_PARSED_ENTITY:
1590 case XML_EXTERNAL_PARAMETER_ENTITY:
1591 input = xmlLoadExternalEntity((char *) entity->URI,
1592 (char *) entity->ExternalID, ctxt);
1593 if (input != NULL)
1594 input->entity = entity;
1595 return(input);
1596 case XML_INTERNAL_GENERAL_ENTITY:
1597 xmlErrInternal(ctxt,
1598 "Internal entity %s without content !\n",
1599 entity->name);
1600 break;
1601 case XML_INTERNAL_PARAMETER_ENTITY:
1602 xmlErrInternal(ctxt,
1603 "Internal parameter entity %s without content !\n",
1604 entity->name);
1605 break;
1606 case XML_INTERNAL_PREDEFINED_ENTITY:
1607 xmlErrInternal(ctxt,
1608 "Predefined entity %s without content !\n",
1609 entity->name);
1610 break;
1611 }
1612 return(NULL);
1613 }
1614 input = xmlNewInputStream(ctxt);
1615 if (input == NULL) {
1616 return(NULL);
1617 }
1618 if (entity->URI != NULL)
1619 input->filename = (char *) xmlStrdup((xmlChar *) entity->URI);
1620 input->base = entity->content;
1621 if (entity->length == 0)
1622 entity->length = xmlStrlen(entity->content);
1623 input->cur = entity->content;
1624 input->length = entity->length;
1625 input->end = &entity->content[input->length];
1626 input->entity = entity;
1627 return(input);
1628 }
1629
1630 /**
1631 * xmlNewStringInputStream:
1632 * @ctxt: an XML parser context
1633 * @buffer: an memory buffer
1634 *
1635 * Create a new input stream based on a memory buffer.
1636 * Returns the new input stream
1637 */
1638 xmlParserInputPtr
xmlNewStringInputStream(xmlParserCtxtPtr ctxt,const xmlChar * buffer)1639 xmlNewStringInputStream(xmlParserCtxtPtr ctxt, const xmlChar *buffer) {
1640 xmlParserInputPtr input;
1641 xmlParserInputBufferPtr buf;
1642
1643 if (buffer == NULL) {
1644 xmlErrInternal(ctxt, "xmlNewStringInputStream string = NULL\n",
1645 NULL);
1646 return(NULL);
1647 }
1648 if (xmlParserDebugEntities)
1649 xmlGenericError(xmlGenericErrorContext,
1650 "new fixed input: %.30s\n", buffer);
1651 buf = xmlParserInputBufferCreateString(buffer);
1652 if (buf == NULL) {
1653 xmlErrMemory(ctxt, NULL);
1654 return(NULL);
1655 }
1656 input = xmlNewInputStream(ctxt);
1657 if (input == NULL) {
1658 xmlErrMemory(ctxt, "couldn't allocate a new input stream\n");
1659 xmlFreeParserInputBuffer(buf);
1660 return(NULL);
1661 }
1662 input->buf = buf;
1663 xmlBufResetInput(input->buf->buffer, input);
1664 return(input);
1665 }
1666
1667 /**
1668 * xmlNewInputFromFile:
1669 * @ctxt: an XML parser context
1670 * @filename: the filename to use as entity
1671 *
1672 * Create a new input stream based on a file or an URL.
1673 *
1674 * Returns the new input stream or NULL in case of error
1675 */
1676 xmlParserInputPtr
xmlNewInputFromFile(xmlParserCtxtPtr ctxt,const char * filename)1677 xmlNewInputFromFile(xmlParserCtxtPtr ctxt, const char *filename) {
1678 xmlParserInputBufferPtr buf;
1679 xmlParserInputPtr inputStream;
1680 char *directory = NULL;
1681 xmlChar *URI = NULL;
1682
1683 if (xmlParserDebugEntities)
1684 xmlGenericError(xmlGenericErrorContext,
1685 "new input from file: %s\n", filename);
1686 if (ctxt == NULL) return(NULL);
1687 buf = xmlParserInputBufferCreateFilename(filename, XML_CHAR_ENCODING_NONE);
1688 if (buf == NULL) {
1689 if (filename == NULL)
1690 __xmlLoaderErr(ctxt,
1691 "failed to load external entity: NULL filename \n",
1692 NULL);
1693 else
1694 __xmlLoaderErr(ctxt, "failed to load external entity \"%s\"\n",
1695 (const char *) filename);
1696 return(NULL);
1697 }
1698
1699 inputStream = xmlNewInputStream(ctxt);
1700 if (inputStream == NULL) {
1701 xmlFreeParserInputBuffer(buf);
1702 return(NULL);
1703 }
1704
1705 inputStream->buf = buf;
1706 inputStream = xmlCheckHTTPInput(ctxt, inputStream);
1707 if (inputStream == NULL)
1708 return(NULL);
1709
1710 if (inputStream->filename == NULL)
1711 URI = xmlStrdup((xmlChar *) filename);
1712 else
1713 URI = xmlStrdup((xmlChar *) inputStream->filename);
1714 directory = xmlParserGetDirectory((const char *) URI);
1715 if (inputStream->filename != NULL) xmlFree((char *)inputStream->filename);
1716 inputStream->filename = (char *) xmlCanonicPath((const xmlChar *) URI);
1717 if (URI != NULL) xmlFree((char *) URI);
1718 inputStream->directory = directory;
1719
1720 xmlBufResetInput(inputStream->buf->buffer, inputStream);
1721 if ((ctxt->directory == NULL) && (directory != NULL))
1722 ctxt->directory = (char *) xmlStrdup((const xmlChar *) directory);
1723 return(inputStream);
1724 }
1725
1726 /************************************************************************
1727 * *
1728 * Commodity functions to handle parser contexts *
1729 * *
1730 ************************************************************************/
1731
1732 /**
1733 * xmlInitSAXParserCtxt:
1734 * @ctxt: XML parser context
1735 * @sax: SAX handlert
1736 * @userData: user data
1737 *
1738 * Initialize a SAX parser context
1739 *
1740 * Returns 0 in case of success and -1 in case of error
1741 */
1742
1743 static int
xmlInitSAXParserCtxt(xmlParserCtxtPtr ctxt,const xmlSAXHandler * sax,void * userData)1744 xmlInitSAXParserCtxt(xmlParserCtxtPtr ctxt, const xmlSAXHandler *sax,
1745 void *userData)
1746 {
1747 xmlParserInputPtr input;
1748
1749 if(ctxt==NULL) {
1750 xmlErrInternal(NULL, "Got NULL parser context\n", NULL);
1751 return(-1);
1752 }
1753
1754 xmlInitParser();
1755
1756 if (ctxt->dict == NULL)
1757 ctxt->dict = xmlDictCreate();
1758 if (ctxt->dict == NULL) {
1759 xmlErrMemory(NULL, "cannot initialize parser context\n");
1760 return(-1);
1761 }
1762 xmlDictSetLimit(ctxt->dict, XML_MAX_DICTIONARY_LIMIT);
1763
1764 if (ctxt->sax == NULL)
1765 ctxt->sax = (xmlSAXHandler *) xmlMalloc(sizeof(xmlSAXHandler));
1766 if (ctxt->sax == NULL) {
1767 xmlErrMemory(NULL, "cannot initialize parser context\n");
1768 return(-1);
1769 }
1770 if (sax == NULL) {
1771 memset(ctxt->sax, 0, sizeof(xmlSAXHandler));
1772 xmlSAXVersion(ctxt->sax, 2);
1773 ctxt->userData = ctxt;
1774 } else {
1775 if (sax->initialized == XML_SAX2_MAGIC) {
1776 memcpy(ctxt->sax, sax, sizeof(xmlSAXHandler));
1777 } else {
1778 memset(ctxt->sax, 0, sizeof(xmlSAXHandler));
1779 memcpy(ctxt->sax, sax, sizeof(xmlSAXHandlerV1));
1780 }
1781 ctxt->userData = userData ? userData : ctxt;
1782 }
1783
1784 ctxt->maxatts = 0;
1785 ctxt->atts = NULL;
1786 /* Allocate the Input stack */
1787 if (ctxt->inputTab == NULL) {
1788 ctxt->inputTab = (xmlParserInputPtr *)
1789 xmlMalloc(5 * sizeof(xmlParserInputPtr));
1790 ctxt->inputMax = 5;
1791 }
1792 if (ctxt->inputTab == NULL) {
1793 xmlErrMemory(NULL, "cannot initialize parser context\n");
1794 ctxt->inputNr = 0;
1795 ctxt->inputMax = 0;
1796 ctxt->input = NULL;
1797 return(-1);
1798 }
1799 while ((input = inputPop(ctxt)) != NULL) { /* Non consuming */
1800 xmlFreeInputStream(input);
1801 }
1802 ctxt->inputNr = 0;
1803 ctxt->input = NULL;
1804
1805 ctxt->version = NULL;
1806 ctxt->encoding = NULL;
1807 ctxt->standalone = -1;
1808 ctxt->hasExternalSubset = 0;
1809 ctxt->hasPErefs = 0;
1810 ctxt->html = 0;
1811 ctxt->external = 0;
1812 ctxt->instate = XML_PARSER_START;
1813 ctxt->token = 0;
1814 ctxt->directory = NULL;
1815
1816 /* Allocate the Node stack */
1817 if (ctxt->nodeTab == NULL) {
1818 ctxt->nodeTab = (xmlNodePtr *) xmlMalloc(10 * sizeof(xmlNodePtr));
1819 ctxt->nodeMax = 10;
1820 }
1821 if (ctxt->nodeTab == NULL) {
1822 xmlErrMemory(NULL, "cannot initialize parser context\n");
1823 ctxt->nodeNr = 0;
1824 ctxt->nodeMax = 0;
1825 ctxt->node = NULL;
1826 ctxt->inputNr = 0;
1827 ctxt->inputMax = 0;
1828 ctxt->input = NULL;
1829 return(-1);
1830 }
1831 ctxt->nodeNr = 0;
1832 ctxt->node = NULL;
1833
1834 /* Allocate the Name stack */
1835 if (ctxt->nameTab == NULL) {
1836 ctxt->nameTab = (const xmlChar **) xmlMalloc(10 * sizeof(xmlChar *));
1837 ctxt->nameMax = 10;
1838 }
1839 if (ctxt->nameTab == NULL) {
1840 xmlErrMemory(NULL, "cannot initialize parser context\n");
1841 ctxt->nodeNr = 0;
1842 ctxt->nodeMax = 0;
1843 ctxt->node = NULL;
1844 ctxt->inputNr = 0;
1845 ctxt->inputMax = 0;
1846 ctxt->input = NULL;
1847 ctxt->nameNr = 0;
1848 ctxt->nameMax = 0;
1849 ctxt->name = NULL;
1850 return(-1);
1851 }
1852 ctxt->nameNr = 0;
1853 ctxt->name = NULL;
1854
1855 /* Allocate the space stack */
1856 if (ctxt->spaceTab == NULL) {
1857 ctxt->spaceTab = (int *) xmlMalloc(10 * sizeof(int));
1858 ctxt->spaceMax = 10;
1859 }
1860 if (ctxt->spaceTab == NULL) {
1861 xmlErrMemory(NULL, "cannot initialize parser context\n");
1862 ctxt->nodeNr = 0;
1863 ctxt->nodeMax = 0;
1864 ctxt->node = NULL;
1865 ctxt->inputNr = 0;
1866 ctxt->inputMax = 0;
1867 ctxt->input = NULL;
1868 ctxt->nameNr = 0;
1869 ctxt->nameMax = 0;
1870 ctxt->name = NULL;
1871 ctxt->spaceNr = 0;
1872 ctxt->spaceMax = 0;
1873 ctxt->space = NULL;
1874 return(-1);
1875 }
1876 ctxt->spaceNr = 1;
1877 ctxt->spaceMax = 10;
1878 ctxt->spaceTab[0] = -1;
1879 ctxt->space = &ctxt->spaceTab[0];
1880 ctxt->myDoc = NULL;
1881 ctxt->wellFormed = 1;
1882 ctxt->nsWellFormed = 1;
1883 ctxt->valid = 1;
1884 ctxt->loadsubset = xmlLoadExtDtdDefaultValue;
1885 if (ctxt->loadsubset) {
1886 ctxt->options |= XML_PARSE_DTDLOAD;
1887 }
1888 ctxt->validate = xmlDoValidityCheckingDefaultValue;
1889 ctxt->pedantic = xmlPedanticParserDefaultValue;
1890 if (ctxt->pedantic) {
1891 ctxt->options |= XML_PARSE_PEDANTIC;
1892 }
1893 ctxt->linenumbers = xmlLineNumbersDefaultValue;
1894 ctxt->keepBlanks = xmlKeepBlanksDefaultValue;
1895 if (ctxt->keepBlanks == 0) {
1896 ctxt->sax->ignorableWhitespace = xmlSAX2IgnorableWhitespace;
1897 ctxt->options |= XML_PARSE_NOBLANKS;
1898 }
1899
1900 ctxt->vctxt.flags = XML_VCTXT_USE_PCTXT;
1901 ctxt->vctxt.userData = ctxt;
1902 ctxt->vctxt.error = xmlParserValidityError;
1903 ctxt->vctxt.warning = xmlParserValidityWarning;
1904 if (ctxt->validate) {
1905 if (xmlGetWarningsDefaultValue == 0)
1906 ctxt->vctxt.warning = NULL;
1907 else
1908 ctxt->vctxt.warning = xmlParserValidityWarning;
1909 ctxt->vctxt.nodeMax = 0;
1910 ctxt->options |= XML_PARSE_DTDVALID;
1911 }
1912 ctxt->replaceEntities = xmlSubstituteEntitiesDefaultValue;
1913 if (ctxt->replaceEntities) {
1914 ctxt->options |= XML_PARSE_NOENT;
1915 }
1916 ctxt->record_info = 0;
1917 ctxt->checkIndex = 0;
1918 ctxt->inSubset = 0;
1919 ctxt->errNo = XML_ERR_OK;
1920 ctxt->depth = 0;
1921 ctxt->catalogs = NULL;
1922 ctxt->sizeentities = 0;
1923 ctxt->sizeentcopy = 0;
1924 ctxt->input_id = 1;
1925 ctxt->maxAmpl = XML_MAX_AMPLIFICATION_DEFAULT;
1926 xmlInitNodeInfoSeq(&ctxt->node_seq);
1927
1928 if (ctxt->nsdb == NULL) {
1929 ctxt->nsdb = xmlParserNsCreate();
1930 if (ctxt->nsdb == NULL) {
1931 xmlErrMemory(ctxt, NULL);
1932 return(-1);
1933 }
1934 }
1935
1936 return(0);
1937 }
1938
1939 /**
1940 * xmlInitParserCtxt:
1941 * @ctxt: an XML parser context
1942 *
1943 * DEPRECATED: Internal function which will be made private in a future
1944 * version.
1945 *
1946 * Initialize a parser context
1947 *
1948 * Returns 0 in case of success and -1 in case of error
1949 */
1950
1951 int
xmlInitParserCtxt(xmlParserCtxtPtr ctxt)1952 xmlInitParserCtxt(xmlParserCtxtPtr ctxt)
1953 {
1954 return(xmlInitSAXParserCtxt(ctxt, NULL, NULL));
1955 }
1956
1957 /**
1958 * xmlFreeParserCtxt:
1959 * @ctxt: an XML parser context
1960 *
1961 * Free all the memory used by a parser context. However the parsed
1962 * document in ctxt->myDoc is not freed.
1963 */
1964
1965 void
xmlFreeParserCtxt(xmlParserCtxtPtr ctxt)1966 xmlFreeParserCtxt(xmlParserCtxtPtr ctxt)
1967 {
1968 xmlParserInputPtr input;
1969
1970 if (ctxt == NULL) return;
1971
1972 while ((input = inputPop(ctxt)) != NULL) { /* Non consuming */
1973 xmlFreeInputStream(input);
1974 }
1975 if (ctxt->spaceTab != NULL) xmlFree(ctxt->spaceTab);
1976 if (ctxt->nameTab != NULL) xmlFree((xmlChar * *)ctxt->nameTab);
1977 if (ctxt->nodeTab != NULL) xmlFree(ctxt->nodeTab);
1978 if (ctxt->nodeInfoTab != NULL) xmlFree(ctxt->nodeInfoTab);
1979 if (ctxt->inputTab != NULL) xmlFree(ctxt->inputTab);
1980 if (ctxt->version != NULL) xmlFree((char *) ctxt->version);
1981 if (ctxt->encoding != NULL) xmlFree((char *) ctxt->encoding);
1982 if (ctxt->extSubURI != NULL) xmlFree((char *) ctxt->extSubURI);
1983 if (ctxt->extSubSystem != NULL) xmlFree((char *) ctxt->extSubSystem);
1984 #ifdef LIBXML_SAX1_ENABLED
1985 if ((ctxt->sax != NULL) &&
1986 (ctxt->sax != (xmlSAXHandlerPtr) &xmlDefaultSAXHandler))
1987 #else
1988 if (ctxt->sax != NULL)
1989 #endif /* LIBXML_SAX1_ENABLED */
1990 xmlFree(ctxt->sax);
1991 if (ctxt->directory != NULL) xmlFree((char *) ctxt->directory);
1992 if (ctxt->vctxt.nodeTab != NULL) xmlFree(ctxt->vctxt.nodeTab);
1993 if (ctxt->atts != NULL) xmlFree((xmlChar * *)ctxt->atts);
1994 if (ctxt->dict != NULL) xmlDictFree(ctxt->dict);
1995 if (ctxt->nsTab != NULL) xmlFree(ctxt->nsTab);
1996 if (ctxt->nsdb != NULL) xmlParserNsFree(ctxt->nsdb);
1997 if (ctxt->attrHash != NULL) xmlFree(ctxt->attrHash);
1998 if (ctxt->pushTab != NULL) xmlFree(ctxt->pushTab);
1999 if (ctxt->attallocs != NULL) xmlFree(ctxt->attallocs);
2000 if (ctxt->attsDefault != NULL)
2001 xmlHashFree(ctxt->attsDefault, xmlHashDefaultDeallocator);
2002 if (ctxt->attsSpecial != NULL)
2003 xmlHashFree(ctxt->attsSpecial, NULL);
2004 if (ctxt->freeElems != NULL) {
2005 xmlNodePtr cur, next;
2006
2007 cur = ctxt->freeElems;
2008 while (cur != NULL) {
2009 next = cur->next;
2010 xmlFree(cur);
2011 cur = next;
2012 }
2013 }
2014 if (ctxt->freeAttrs != NULL) {
2015 xmlAttrPtr cur, next;
2016
2017 cur = ctxt->freeAttrs;
2018 while (cur != NULL) {
2019 next = cur->next;
2020 xmlFree(cur);
2021 cur = next;
2022 }
2023 }
2024 /*
2025 * cleanup the error strings
2026 */
2027 if (ctxt->lastError.message != NULL)
2028 xmlFree(ctxt->lastError.message);
2029 if (ctxt->lastError.file != NULL)
2030 xmlFree(ctxt->lastError.file);
2031 if (ctxt->lastError.str1 != NULL)
2032 xmlFree(ctxt->lastError.str1);
2033 if (ctxt->lastError.str2 != NULL)
2034 xmlFree(ctxt->lastError.str2);
2035 if (ctxt->lastError.str3 != NULL)
2036 xmlFree(ctxt->lastError.str3);
2037
2038 #ifdef LIBXML_CATALOG_ENABLED
2039 if (ctxt->catalogs != NULL)
2040 xmlCatalogFreeLocal(ctxt->catalogs);
2041 #endif
2042 xmlFree(ctxt);
2043 }
2044
2045 /**
2046 * xmlNewParserCtxt:
2047 *
2048 * Allocate and initialize a new parser context.
2049 *
2050 * Returns the xmlParserCtxtPtr or NULL
2051 */
2052
2053 xmlParserCtxtPtr
xmlNewParserCtxt(void)2054 xmlNewParserCtxt(void)
2055 {
2056 return(xmlNewSAXParserCtxt(NULL, NULL));
2057 }
2058
2059 /**
2060 * xmlNewSAXParserCtxt:
2061 * @sax: SAX handler
2062 * @userData: user data
2063 *
2064 * Allocate and initialize a new SAX parser context. If userData is NULL,
2065 * the parser context will be passed as user data.
2066 *
2067 * Returns the xmlParserCtxtPtr or NULL if memory allocation failed.
2068 */
2069
2070 xmlParserCtxtPtr
xmlNewSAXParserCtxt(const xmlSAXHandler * sax,void * userData)2071 xmlNewSAXParserCtxt(const xmlSAXHandler *sax, void *userData)
2072 {
2073 xmlParserCtxtPtr ctxt;
2074
2075 ctxt = (xmlParserCtxtPtr) xmlMalloc(sizeof(xmlParserCtxt));
2076 if (ctxt == NULL) {
2077 xmlErrMemory(NULL, "cannot allocate parser context\n");
2078 return(NULL);
2079 }
2080 memset(ctxt, 0, sizeof(xmlParserCtxt));
2081 if (xmlInitSAXParserCtxt(ctxt, sax, userData) < 0) {
2082 xmlFreeParserCtxt(ctxt);
2083 return(NULL);
2084 }
2085 return(ctxt);
2086 }
2087
2088 /************************************************************************
2089 * *
2090 * Handling of node information *
2091 * *
2092 ************************************************************************/
2093
2094 /**
2095 * xmlClearParserCtxt:
2096 * @ctxt: an XML parser context
2097 *
2098 * Clear (release owned resources) and reinitialize a parser context
2099 */
2100
2101 void
xmlClearParserCtxt(xmlParserCtxtPtr ctxt)2102 xmlClearParserCtxt(xmlParserCtxtPtr ctxt)
2103 {
2104 if (ctxt==NULL)
2105 return;
2106 xmlClearNodeInfoSeq(&ctxt->node_seq);
2107 xmlCtxtReset(ctxt);
2108 }
2109
2110
2111 /**
2112 * xmlParserFindNodeInfo:
2113 * @ctx: an XML parser context
2114 * @node: an XML node within the tree
2115 *
2116 * DEPRECATED: Don't use.
2117 *
2118 * Find the parser node info struct for a given node
2119 *
2120 * Returns an xmlParserNodeInfo block pointer or NULL
2121 */
2122 const xmlParserNodeInfo *
xmlParserFindNodeInfo(const xmlParserCtxtPtr ctx,const xmlNodePtr node)2123 xmlParserFindNodeInfo(const xmlParserCtxtPtr ctx, const xmlNodePtr node)
2124 {
2125 unsigned long pos;
2126
2127 if ((ctx == NULL) || (node == NULL))
2128 return (NULL);
2129 /* Find position where node should be at */
2130 pos = xmlParserFindNodeInfoIndex(&ctx->node_seq, node);
2131 if (pos < ctx->node_seq.length
2132 && ctx->node_seq.buffer[pos].node == node)
2133 return &ctx->node_seq.buffer[pos];
2134 else
2135 return NULL;
2136 }
2137
2138
2139 /**
2140 * xmlInitNodeInfoSeq:
2141 * @seq: a node info sequence pointer
2142 *
2143 * DEPRECATED: Don't use.
2144 *
2145 * -- Initialize (set to initial state) node info sequence
2146 */
2147 void
xmlInitNodeInfoSeq(xmlParserNodeInfoSeqPtr seq)2148 xmlInitNodeInfoSeq(xmlParserNodeInfoSeqPtr seq)
2149 {
2150 if (seq == NULL)
2151 return;
2152 seq->length = 0;
2153 seq->maximum = 0;
2154 seq->buffer = NULL;
2155 }
2156
2157 /**
2158 * xmlClearNodeInfoSeq:
2159 * @seq: a node info sequence pointer
2160 *
2161 * DEPRECATED: Don't use.
2162 *
2163 * -- Clear (release memory and reinitialize) node
2164 * info sequence
2165 */
2166 void
xmlClearNodeInfoSeq(xmlParserNodeInfoSeqPtr seq)2167 xmlClearNodeInfoSeq(xmlParserNodeInfoSeqPtr seq)
2168 {
2169 if (seq == NULL)
2170 return;
2171 if (seq->buffer != NULL)
2172 xmlFree(seq->buffer);
2173 xmlInitNodeInfoSeq(seq);
2174 }
2175
2176 /**
2177 * xmlParserFindNodeInfoIndex:
2178 * @seq: a node info sequence pointer
2179 * @node: an XML node pointer
2180 *
2181 * DEPRECATED: Don't use.
2182 *
2183 * xmlParserFindNodeInfoIndex : Find the index that the info record for
2184 * the given node is or should be at in a sorted sequence
2185 *
2186 * Returns a long indicating the position of the record
2187 */
2188 unsigned long
xmlParserFindNodeInfoIndex(const xmlParserNodeInfoSeqPtr seq,const xmlNodePtr node)2189 xmlParserFindNodeInfoIndex(const xmlParserNodeInfoSeqPtr seq,
2190 const xmlNodePtr node)
2191 {
2192 unsigned long upper, lower, middle;
2193 int found = 0;
2194
2195 if ((seq == NULL) || (node == NULL))
2196 return ((unsigned long) -1);
2197
2198 /* Do a binary search for the key */
2199 lower = 1;
2200 upper = seq->length;
2201 middle = 0;
2202 while (lower <= upper && !found) {
2203 middle = lower + (upper - lower) / 2;
2204 if (node == seq->buffer[middle - 1].node)
2205 found = 1;
2206 else if (node < seq->buffer[middle - 1].node)
2207 upper = middle - 1;
2208 else
2209 lower = middle + 1;
2210 }
2211
2212 /* Return position */
2213 if (middle == 0 || seq->buffer[middle - 1].node < node)
2214 return middle;
2215 else
2216 return middle - 1;
2217 }
2218
2219
2220 /**
2221 * xmlParserAddNodeInfo:
2222 * @ctxt: an XML parser context
2223 * @info: a node info sequence pointer
2224 *
2225 * DEPRECATED: Don't use.
2226 *
2227 * Insert node info record into the sorted sequence
2228 */
2229 void
xmlParserAddNodeInfo(xmlParserCtxtPtr ctxt,const xmlParserNodeInfoPtr info)2230 xmlParserAddNodeInfo(xmlParserCtxtPtr ctxt,
2231 const xmlParserNodeInfoPtr info)
2232 {
2233 unsigned long pos;
2234
2235 if ((ctxt == NULL) || (info == NULL)) return;
2236
2237 /* Find pos and check to see if node is already in the sequence */
2238 pos = xmlParserFindNodeInfoIndex(&ctxt->node_seq, (xmlNodePtr)
2239 info->node);
2240
2241 if ((pos < ctxt->node_seq.length) &&
2242 (ctxt->node_seq.buffer != NULL) &&
2243 (ctxt->node_seq.buffer[pos].node == info->node)) {
2244 ctxt->node_seq.buffer[pos] = *info;
2245 }
2246
2247 /* Otherwise, we need to add new node to buffer */
2248 else {
2249 if ((ctxt->node_seq.length + 1 > ctxt->node_seq.maximum) ||
2250 (ctxt->node_seq.buffer == NULL)) {
2251 xmlParserNodeInfo *tmp_buffer;
2252 unsigned int byte_size;
2253
2254 if (ctxt->node_seq.maximum == 0)
2255 ctxt->node_seq.maximum = 2;
2256 byte_size = (sizeof(*ctxt->node_seq.buffer) *
2257 (2 * ctxt->node_seq.maximum));
2258
2259 if (ctxt->node_seq.buffer == NULL)
2260 tmp_buffer = (xmlParserNodeInfo *) xmlMalloc(byte_size);
2261 else
2262 tmp_buffer =
2263 (xmlParserNodeInfo *) xmlRealloc(ctxt->node_seq.buffer,
2264 byte_size);
2265
2266 if (tmp_buffer == NULL) {
2267 xmlErrMemory(ctxt, "failed to allocate buffer\n");
2268 return;
2269 }
2270 ctxt->node_seq.buffer = tmp_buffer;
2271 ctxt->node_seq.maximum *= 2;
2272 }
2273
2274 /* If position is not at end, move elements out of the way */
2275 if (pos != ctxt->node_seq.length) {
2276 unsigned long i;
2277
2278 for (i = ctxt->node_seq.length; i > pos; i--)
2279 ctxt->node_seq.buffer[i] = ctxt->node_seq.buffer[i - 1];
2280 }
2281
2282 /* Copy element and increase length */
2283 ctxt->node_seq.buffer[pos] = *info;
2284 ctxt->node_seq.length++;
2285 }
2286 }
2287
2288 /************************************************************************
2289 * *
2290 * Defaults settings *
2291 * *
2292 ************************************************************************/
2293 /**
2294 * xmlPedanticParserDefault:
2295 * @val: int 0 or 1
2296 *
2297 * DEPRECATED: Use the modern options API with XML_PARSE_PEDANTIC.
2298 *
2299 * Set and return the previous value for enabling pedantic warnings.
2300 *
2301 * Returns the last value for 0 for no substitution, 1 for substitution.
2302 */
2303
2304 int
xmlPedanticParserDefault(int val)2305 xmlPedanticParserDefault(int val) {
2306 int old = xmlPedanticParserDefaultValue;
2307
2308 xmlPedanticParserDefaultValue = val;
2309 return(old);
2310 }
2311
2312 /**
2313 * xmlLineNumbersDefault:
2314 * @val: int 0 or 1
2315 *
2316 * DEPRECATED: The modern options API always enables line numbers.
2317 *
2318 * Set and return the previous value for enabling line numbers in elements
2319 * contents. This may break on old application and is turned off by default.
2320 *
2321 * Returns the last value for 0 for no substitution, 1 for substitution.
2322 */
2323
2324 int
xmlLineNumbersDefault(int val)2325 xmlLineNumbersDefault(int val) {
2326 int old = xmlLineNumbersDefaultValue;
2327
2328 xmlLineNumbersDefaultValue = val;
2329 return(old);
2330 }
2331
2332 /**
2333 * xmlSubstituteEntitiesDefault:
2334 * @val: int 0 or 1
2335 *
2336 * DEPRECATED: Use the modern options API with XML_PARSE_NOENT.
2337 *
2338 * Set and return the previous value for default entity support.
2339 * Initially the parser always keep entity references instead of substituting
2340 * entity values in the output. This function has to be used to change the
2341 * default parser behavior
2342 * SAX::substituteEntities() has to be used for changing that on a file by
2343 * file basis.
2344 *
2345 * Returns the last value for 0 for no substitution, 1 for substitution.
2346 */
2347
2348 int
xmlSubstituteEntitiesDefault(int val)2349 xmlSubstituteEntitiesDefault(int val) {
2350 int old = xmlSubstituteEntitiesDefaultValue;
2351
2352 xmlSubstituteEntitiesDefaultValue = val;
2353 return(old);
2354 }
2355
2356 /**
2357 * xmlKeepBlanksDefault:
2358 * @val: int 0 or 1
2359 *
2360 * DEPRECATED: Use the modern options API with XML_PARSE_NOBLANKS.
2361 *
2362 * Set and return the previous value for default blanks text nodes support.
2363 * The 1.x version of the parser used an heuristic to try to detect
2364 * ignorable white spaces. As a result the SAX callback was generating
2365 * xmlSAX2IgnorableWhitespace() callbacks instead of characters() one, and when
2366 * using the DOM output text nodes containing those blanks were not generated.
2367 * The 2.x and later version will switch to the XML standard way and
2368 * ignorableWhitespace() are only generated when running the parser in
2369 * validating mode and when the current element doesn't allow CDATA or
2370 * mixed content.
2371 * This function is provided as a way to force the standard behavior
2372 * on 1.X libs and to switch back to the old mode for compatibility when
2373 * running 1.X client code on 2.X . Upgrade of 1.X code should be done
2374 * by using xmlIsBlankNode() commodity function to detect the "empty"
2375 * nodes generated.
2376 * This value also affect autogeneration of indentation when saving code
2377 * if blanks sections are kept, indentation is not generated.
2378 *
2379 * Returns the last value for 0 for no substitution, 1 for substitution.
2380 */
2381
2382 int
xmlKeepBlanksDefault(int val)2383 xmlKeepBlanksDefault(int val) {
2384 int old = xmlKeepBlanksDefaultValue;
2385
2386 xmlKeepBlanksDefaultValue = val;
2387 #ifdef LIBXML_OUTPUT_ENABLED
2388 if (!val)
2389 xmlIndentTreeOutput = 1;
2390 #endif
2391 return(old);
2392 }
2393
2394