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
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
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
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)
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
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)
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
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
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
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
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
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
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
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
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
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 /*
853 * TODO: This function shouldn't change the 'cur' pointer
854 * as side effect, but the NEXTL macro in parser.c relies
855 * on this behavior when incrementing line numbers.
856 */
857 if (cur[1] == '\n')
858 ctxt->input->cur++;
859 *len = 1;
860 c = '\n';
861 } else if (c == 0) {
862 if (ctxt->input->cur >= ctxt->input->end) {
863 *len = 0;
864 } else {
865 *len = 1;
866 /*
867 * TODO: Null bytes should be handled by callers,
868 * but this can be tricky.
869 */
870 xmlErrEncodingInt(ctxt, XML_ERR_INVALID_CHAR,
871 "Char 0x0 out of allowed range\n", c);
872 }
873 } else {
874 *len = 1;
875 }
876 } else {
877 *len = 1;
878 }
879
880 return(c);
881 } else {
882 int val;
883
884 if (avail < 2)
885 goto incomplete_sequence;
886 if ((cur[1] & 0xc0) != 0x80)
887 goto encoding_error;
888
889 if (c < 0xe0) {
890 /* 2-byte code */
891 if (c < 0xc2)
892 goto encoding_error;
893 val = (c & 0x1f) << 6;
894 val |= cur[1] & 0x3f;
895 *len = 2;
896 } else {
897 if (avail < 3)
898 goto incomplete_sequence;
899 if ((cur[2] & 0xc0) != 0x80)
900 goto encoding_error;
901
902 if (c < 0xf0) {
903 /* 3-byte code */
904 val = (c & 0xf) << 12;
905 val |= (cur[1] & 0x3f) << 6;
906 val |= cur[2] & 0x3f;
907 if ((val < 0x800) || ((val >= 0xd800) && (val < 0xe000)))
908 goto encoding_error;
909 *len = 3;
910 } else {
911 if (avail < 4)
912 goto incomplete_sequence;
913 if ((cur[3] & 0xc0) != 0x80)
914 goto encoding_error;
915
916 /* 4-byte code */
917 val = (c & 0x0f) << 18;
918 val |= (cur[1] & 0x3f) << 12;
919 val |= (cur[2] & 0x3f) << 6;
920 val |= cur[3] & 0x3f;
921 if ((val < 0x10000) || (val >= 0x110000))
922 goto encoding_error;
923 *len = 4;
924 }
925 }
926
927 return(val);
928 }
929
930 encoding_error:
931 /* Only report the first error */
932 if ((ctxt->input->flags & XML_INPUT_ENCODING_ERROR) == 0) {
933 if (ctxt->input->end - ctxt->input->cur < 4) {
934 __xmlErrEncoding(ctxt, XML_ERR_INVALID_CHAR,
935 "Input is not proper UTF-8, indicate encoding !\n",
936 NULL, NULL);
937 } else {
938 char buffer[150];
939
940 snprintf(&buffer[0], 149, "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n",
941 ctxt->input->cur[0], ctxt->input->cur[1],
942 ctxt->input->cur[2], ctxt->input->cur[3]);
943 __xmlErrEncoding(ctxt, XML_ERR_INVALID_CHAR,
944 "Input is not proper UTF-8, indicate encoding !\n%s",
945 BAD_CAST buffer, NULL);
946 }
947 ctxt->input->flags |= XML_INPUT_ENCODING_ERROR;
948 }
949 *len = 1;
950 return(0xFFFD); /* U+FFFD Replacement Character */
951
952 incomplete_sequence:
953 /*
954 * An encoding problem may arise from a truncated input buffer
955 * splitting a character in the middle. In that case do not raise
956 * an error but return 0. This should only happen when push parsing
957 * char data.
958 */
959 *len = 0;
960 return(0);
961 }
962
963 /**
964 * xmlStringCurrentChar:
965 * @ctxt: the XML parser context
966 * @cur: pointer to the beginning of the char
967 * @len: pointer to the length of the char read
968 *
969 * DEPRECATED: Internal function, do not use.
970 *
971 * The current char value, if using UTF-8 this may actually span multiple
972 * bytes in the input buffer.
973 *
974 * Returns the current char value and its length
975 */
976
977 int
978 xmlStringCurrentChar(xmlParserCtxtPtr ctxt ATTRIBUTE_UNUSED,
979 const xmlChar *cur, int *len) {
980 int c;
981
982 if ((cur == NULL) || (len == NULL))
983 return(0);
984
985 /* cur is zero-terminated, so we can lie about its length. */
986 *len = 4;
987 c = xmlGetUTF8Char(cur, len);
988
989 return((c < 0) ? 0 : c);
990 }
991
992 /**
993 * xmlCopyCharMultiByte:
994 * @out: pointer to an array of xmlChar
995 * @val: the char value
996 *
997 * append the char value in the array
998 *
999 * Returns the number of xmlChar written
1000 */
1001 int
1002 xmlCopyCharMultiByte(xmlChar *out, int val) {
1003 if ((out == NULL) || (val < 0)) return(0);
1004 /*
1005 * We are supposed to handle UTF8, check it's valid
1006 * From rfc2044: encoding of the Unicode values on UTF-8:
1007 *
1008 * UCS-4 range (hex.) UTF-8 octet sequence (binary)
1009 * 0000 0000-0000 007F 0xxxxxxx
1010 * 0000 0080-0000 07FF 110xxxxx 10xxxxxx
1011 * 0000 0800-0000 FFFF 1110xxxx 10xxxxxx 10xxxxxx
1012 */
1013 if (val >= 0x80) {
1014 xmlChar *savedout = out;
1015 int bits;
1016 if (val < 0x800) { *out++= (val >> 6) | 0xC0; bits= 0; }
1017 else if (val < 0x10000) { *out++= (val >> 12) | 0xE0; bits= 6;}
1018 else if (val < 0x110000) { *out++= (val >> 18) | 0xF0; bits= 12; }
1019 else {
1020 xmlErrEncodingInt(NULL, XML_ERR_INVALID_CHAR,
1021 "Internal error, xmlCopyCharMultiByte 0x%X out of bound\n",
1022 val);
1023 return(0);
1024 }
1025 for ( ; bits >= 0; bits-= 6)
1026 *out++= ((val >> bits) & 0x3F) | 0x80 ;
1027 return (out - savedout);
1028 }
1029 *out = val;
1030 return 1;
1031 }
1032
1033 /**
1034 * xmlCopyChar:
1035 * @len: Ignored, compatibility
1036 * @out: pointer to an array of xmlChar
1037 * @val: the char value
1038 *
1039 * append the char value in the array
1040 *
1041 * Returns the number of xmlChar written
1042 */
1043
1044 int
1045 xmlCopyChar(int len ATTRIBUTE_UNUSED, xmlChar *out, int val) {
1046 if ((out == NULL) || (val < 0)) return(0);
1047 /* the len parameter is ignored */
1048 if (val >= 0x80) {
1049 return(xmlCopyCharMultiByte (out, val));
1050 }
1051 *out = val;
1052 return 1;
1053 }
1054
1055 /************************************************************************
1056 * *
1057 * Commodity functions to switch encodings *
1058 * *
1059 ************************************************************************/
1060
1061 static xmlCharEncodingHandlerPtr
1062 xmlDetectEBCDIC(xmlParserInputPtr input) {
1063 xmlChar out[200];
1064 xmlCharEncodingHandlerPtr handler;
1065 int inlen, outlen, res, i;
1066
1067 /*
1068 * To detect the EBCDIC code page, we convert the first 200 bytes
1069 * to EBCDIC-US and try to find the encoding declaration.
1070 */
1071 handler = xmlGetCharEncodingHandler(XML_CHAR_ENCODING_EBCDIC);
1072 if (handler == NULL)
1073 return(NULL);
1074 outlen = sizeof(out) - 1;
1075 inlen = input->end - input->cur;
1076 res = xmlEncInputChunk(handler, out, &outlen, input->cur, &inlen);
1077 if (res < 0)
1078 return(handler);
1079 out[outlen] = 0;
1080
1081 for (i = 0; i < outlen; i++) {
1082 if (out[i] == '>')
1083 break;
1084 if ((out[i] == 'e') &&
1085 (xmlStrncmp(out + i, BAD_CAST "encoding", 8) == 0)) {
1086 int start, cur, quote;
1087
1088 i += 8;
1089 while (IS_BLANK_CH(out[i]))
1090 i += 1;
1091 if (out[i++] != '=')
1092 break;
1093 while (IS_BLANK_CH(out[i]))
1094 i += 1;
1095 quote = out[i++];
1096 if ((quote != '\'') && (quote != '"'))
1097 break;
1098 start = i;
1099 cur = out[i];
1100 while (((cur >= 'a') && (cur <= 'z')) ||
1101 ((cur >= 'A') && (cur <= 'Z')) ||
1102 ((cur >= '0') && (cur <= '9')) ||
1103 (cur == '.') || (cur == '_') ||
1104 (cur == '-'))
1105 cur = out[++i];
1106 if (cur != quote)
1107 break;
1108 out[i] = 0;
1109 xmlCharEncCloseFunc(handler);
1110 return(xmlFindCharEncodingHandler((char *) out + start));
1111 }
1112 }
1113
1114 /*
1115 * ICU handlers are stateful, so we have to recreate them.
1116 */
1117 xmlCharEncCloseFunc(handler);
1118 return(xmlGetCharEncodingHandler(XML_CHAR_ENCODING_EBCDIC));
1119 }
1120
1121 /**
1122 * xmlSwitchEncoding:
1123 * @ctxt: the parser context
1124 * @enc: the encoding value (number)
1125 *
1126 * Use encoding specified by enum to decode input data.
1127 *
1128 * This function can be used to enforce the encoding of chunks passed
1129 * to xmlParseChunk.
1130 *
1131 * Returns 0 in case of success, -1 otherwise
1132 */
1133 int
1134 xmlSwitchEncoding(xmlParserCtxtPtr ctxt, xmlCharEncoding enc)
1135 {
1136 xmlCharEncodingHandlerPtr handler = NULL;
1137 int check = 1;
1138 int ret;
1139
1140 if ((ctxt == NULL) || (ctxt->input == NULL))
1141 return(-1);
1142
1143 switch (enc) {
1144 case XML_CHAR_ENCODING_NONE:
1145 case XML_CHAR_ENCODING_UTF8:
1146 case XML_CHAR_ENCODING_ASCII:
1147 check = 0;
1148 break;
1149 case XML_CHAR_ENCODING_EBCDIC:
1150 handler = xmlDetectEBCDIC(ctxt->input);
1151 break;
1152 default:
1153 handler = xmlGetCharEncodingHandler(enc);
1154 break;
1155 }
1156
1157 if ((check) && (handler == NULL)) {
1158 const char *name = xmlGetCharEncodingName(enc);
1159
1160 __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
1161 "encoding not supported: %s\n",
1162 BAD_CAST (name ? name : "<null>"), NULL);
1163 /*
1164 * TODO: We could recover from errors in external entities
1165 * if we didn't stop the parser. But most callers of this
1166 * function don't check the return value.
1167 */
1168 xmlStopParser(ctxt);
1169 return(-1);
1170 }
1171
1172 ret = xmlSwitchInputEncoding(ctxt, ctxt->input, handler);
1173
1174 if ((ret >= 0) && (enc == XML_CHAR_ENCODING_NONE)) {
1175 ctxt->input->flags &= ~XML_INPUT_HAS_ENCODING;
1176 }
1177
1178 return(ret);
1179 }
1180
1181 /**
1182 * xmlSwitchInputEncoding:
1183 * @ctxt: the parser context
1184 * @input: the input stream
1185 * @handler: the encoding handler
1186 *
1187 * DEPRECATED: Internal function, don't use.
1188 *
1189 * Use encoding handler to decode input data.
1190 *
1191 * Returns 0 in case of success, -1 otherwise
1192 */
1193 int
1194 xmlSwitchInputEncoding(xmlParserCtxtPtr ctxt, xmlParserInputPtr input,
1195 xmlCharEncodingHandlerPtr handler)
1196 {
1197 int nbchars;
1198 xmlParserInputBufferPtr in;
1199
1200 if ((input == NULL) || (input->buf == NULL)) {
1201 xmlCharEncCloseFunc(handler);
1202 return (-1);
1203 }
1204 in = input->buf;
1205
1206 input->flags |= XML_INPUT_HAS_ENCODING;
1207
1208 /*
1209 * UTF-8 requires no encoding handler.
1210 */
1211 if ((handler != NULL) &&
1212 (xmlStrcasecmp(BAD_CAST handler->name, BAD_CAST "UTF-8") == 0)) {
1213 xmlCharEncCloseFunc(handler);
1214 handler = NULL;
1215 }
1216
1217 if (in->encoder == handler)
1218 return (0);
1219
1220 if (in->encoder != NULL) {
1221 /*
1222 * Switching encodings during parsing is a really bad idea,
1223 * but Chromium can switch between ISO-8859-1 and UTF-16 before
1224 * separate calls to xmlParseChunk.
1225 *
1226 * TODO: We should check whether the "raw" input buffer is empty and
1227 * convert the old content using the old encoder.
1228 */
1229
1230 xmlCharEncCloseFunc(in->encoder);
1231 in->encoder = handler;
1232 return (0);
1233 }
1234
1235 in->encoder = handler;
1236
1237 /*
1238 * Is there already some content down the pipe to convert ?
1239 */
1240 if (xmlBufIsEmpty(in->buffer) == 0) {
1241 size_t processed;
1242
1243 /*
1244 * Shrink the current input buffer.
1245 * Move it as the raw buffer and create a new input buffer
1246 */
1247 processed = input->cur - input->base;
1248 xmlBufShrink(in->buffer, processed);
1249 input->consumed += processed;
1250 in->raw = in->buffer;
1251 in->buffer = xmlBufCreate();
1252 in->rawconsumed = processed;
1253
1254 nbchars = xmlCharEncInput(in);
1255 xmlBufResetInput(in->buffer, input);
1256 if (nbchars < 0) {
1257 /* TODO: This could be an out of memory or an encoding error. */
1258 xmlErrInternal(ctxt,
1259 "switching encoding: encoder error\n",
1260 NULL);
1261 xmlHaltParser(ctxt);
1262 return (-1);
1263 }
1264 }
1265 return (0);
1266 }
1267
1268 /**
1269 * xmlSwitchToEncoding:
1270 * @ctxt: the parser context
1271 * @handler: the encoding handler
1272 *
1273 * Use encoding handler to decode input data.
1274 *
1275 * This function can be used to enforce the encoding of chunks passed
1276 * to xmlParseChunk.
1277 *
1278 * Returns 0 in case of success, -1 otherwise
1279 */
1280 int
1281 xmlSwitchToEncoding(xmlParserCtxtPtr ctxt, xmlCharEncodingHandlerPtr handler)
1282 {
1283 if (ctxt == NULL)
1284 return(-1);
1285 return(xmlSwitchInputEncoding(ctxt, ctxt->input, handler));
1286 }
1287
1288 /**
1289 * xmlDetectEncoding:
1290 * @ctxt: the parser context
1291 *
1292 * Handle optional BOM, detect and switch to encoding.
1293 *
1294 * Assumes that there are at least four bytes in the input buffer.
1295 */
1296 void
1297 xmlDetectEncoding(xmlParserCtxtPtr ctxt) {
1298 const xmlChar *in;
1299 xmlCharEncoding enc;
1300 int bomSize;
1301 int autoFlag = 0;
1302
1303 if (xmlParserGrow(ctxt) < 0)
1304 return;
1305 in = ctxt->input->cur;
1306 if (ctxt->input->end - in < 4)
1307 return;
1308
1309 if (ctxt->input->flags & XML_INPUT_HAS_ENCODING) {
1310 /*
1311 * If the encoding was already set, only skip the BOM which was
1312 * possibly decoded to UTF-8.
1313 */
1314 if ((in[0] == 0xEF) && (in[1] == 0xBB) && (in[2] == 0xBF)) {
1315 ctxt->input->cur += 3;
1316 }
1317
1318 return;
1319 }
1320
1321 enc = XML_CHAR_ENCODING_NONE;
1322 bomSize = 0;
1323
1324 switch (in[0]) {
1325 case 0x00:
1326 if ((in[1] == 0x00) && (in[2] == 0x00) && (in[3] == 0x3C)) {
1327 enc = XML_CHAR_ENCODING_UCS4BE;
1328 autoFlag = XML_INPUT_AUTO_OTHER;
1329 } else if ((in[1] == 0x3C) && (in[2] == 0x00) && (in[3] == 0x3F)) {
1330 enc = XML_CHAR_ENCODING_UTF16BE;
1331 autoFlag = XML_INPUT_AUTO_UTF16BE;
1332 }
1333 break;
1334
1335 case 0x3C:
1336 if (in[1] == 0x00) {
1337 if ((in[2] == 0x00) && (in[3] == 0x00)) {
1338 enc = XML_CHAR_ENCODING_UCS4LE;
1339 autoFlag = XML_INPUT_AUTO_OTHER;
1340 } else if ((in[2] == 0x3F) && (in[3] == 0x00)) {
1341 enc = XML_CHAR_ENCODING_UTF16LE;
1342 autoFlag = XML_INPUT_AUTO_UTF16LE;
1343 }
1344 }
1345 break;
1346
1347 case 0x4C:
1348 if ((in[1] == 0x6F) && (in[2] == 0xA7) && (in[3] == 0x94)) {
1349 enc = XML_CHAR_ENCODING_EBCDIC;
1350 autoFlag = XML_INPUT_AUTO_OTHER;
1351 }
1352 break;
1353
1354 case 0xEF:
1355 if ((in[1] == 0xBB) && (in[2] == 0xBF)) {
1356 enc = XML_CHAR_ENCODING_UTF8;
1357 autoFlag = XML_INPUT_AUTO_UTF8;
1358 bomSize = 3;
1359 }
1360 break;
1361
1362 case 0xFE:
1363 if (in[1] == 0xFF) {
1364 enc = XML_CHAR_ENCODING_UTF16BE;
1365 autoFlag = XML_INPUT_AUTO_UTF16BE;
1366 bomSize = 2;
1367 }
1368 break;
1369
1370 case 0xFF:
1371 if (in[1] == 0xFE) {
1372 enc = XML_CHAR_ENCODING_UTF16LE;
1373 autoFlag = XML_INPUT_AUTO_UTF16LE;
1374 bomSize = 2;
1375 }
1376 break;
1377 }
1378
1379 if (bomSize > 0) {
1380 ctxt->input->cur += bomSize;
1381 }
1382
1383 if (enc != XML_CHAR_ENCODING_NONE) {
1384 ctxt->input->flags |= autoFlag;
1385 xmlSwitchEncoding(ctxt, enc);
1386 }
1387 }
1388
1389 /**
1390 * xmlSetDeclaredEncoding:
1391 * @ctxt: the parser context
1392 * @encoding: declared encoding
1393 *
1394 * Set the encoding from a declaration in the document.
1395 *
1396 * If no encoding was set yet, switch the encoding. Otherwise, only warn
1397 * about encoding mismatches.
1398 *
1399 * Takes ownership of 'encoding'.
1400 */
1401 void
1402 xmlSetDeclaredEncoding(xmlParserCtxtPtr ctxt, xmlChar *encoding) {
1403 if (ctxt->encoding != NULL)
1404 xmlFree((xmlChar *) ctxt->encoding);
1405 ctxt->encoding = encoding;
1406
1407 if (((ctxt->input->flags & XML_INPUT_HAS_ENCODING) == 0) &&
1408 ((ctxt->options & XML_PARSE_IGNORE_ENC) == 0)) {
1409 xmlCharEncodingHandlerPtr handler;
1410
1411 handler = xmlFindCharEncodingHandler((const char *) encoding);
1412 if (handler == NULL) {
1413 __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
1414 "Unsupported encoding: %s\n",
1415 encoding, NULL);
1416 return;
1417 }
1418
1419 xmlSwitchToEncoding(ctxt, handler);
1420 ctxt->input->flags |= XML_INPUT_USES_ENC_DECL;
1421 } else if (ctxt->input->flags & XML_INPUT_AUTO_ENCODING) {
1422 static const char *allowedUTF8[] = {
1423 "UTF-8", "UTF8", NULL
1424 };
1425 static const char *allowedUTF16LE[] = {
1426 "UTF-16", "UTF-16LE", "UTF16", NULL
1427 };
1428 static const char *allowedUTF16BE[] = {
1429 "UTF-16", "UTF-16BE", "UTF16", NULL
1430 };
1431 const char **allowed = NULL;
1432 const char *autoEnc = NULL;
1433
1434 switch (ctxt->input->flags & XML_INPUT_AUTO_ENCODING) {
1435 case XML_INPUT_AUTO_UTF8:
1436 allowed = allowedUTF8;
1437 autoEnc = "UTF-8";
1438 break;
1439 case XML_INPUT_AUTO_UTF16LE:
1440 allowed = allowedUTF16LE;
1441 autoEnc = "UTF-16LE";
1442 break;
1443 case XML_INPUT_AUTO_UTF16BE:
1444 allowed = allowedUTF16BE;
1445 autoEnc = "UTF-16BE";
1446 break;
1447 }
1448
1449 if (allowed != NULL) {
1450 const char **p;
1451 int match = 0;
1452
1453 for (p = allowed; *p != NULL; p++) {
1454 if (xmlStrcasecmp(encoding, BAD_CAST *p) == 0) {
1455 match = 1;
1456 break;
1457 }
1458 }
1459
1460 if (match == 0) {
1461 xmlWarningMsg(ctxt, XML_WAR_ENCODING_MISMATCH,
1462 "Encoding '%s' doesn't match "
1463 "auto-detected '%s'\n",
1464 encoding, BAD_CAST autoEnc);
1465 }
1466 }
1467 }
1468 }
1469
1470 /************************************************************************
1471 * *
1472 * Commodity functions to handle entities processing *
1473 * *
1474 ************************************************************************/
1475
1476 /**
1477 * xmlFreeInputStream:
1478 * @input: an xmlParserInputPtr
1479 *
1480 * Free up an input stream.
1481 */
1482 void
1483 xmlFreeInputStream(xmlParserInputPtr input) {
1484 if (input == NULL) return;
1485
1486 if (input->filename != NULL) xmlFree((char *) input->filename);
1487 if (input->directory != NULL) xmlFree((char *) input->directory);
1488 if (input->version != NULL) xmlFree((char *) input->version);
1489 if ((input->free != NULL) && (input->base != NULL))
1490 input->free((xmlChar *) input->base);
1491 if (input->buf != NULL)
1492 xmlFreeParserInputBuffer(input->buf);
1493 xmlFree(input);
1494 }
1495
1496 /**
1497 * xmlNewInputStream:
1498 * @ctxt: an XML parser context
1499 *
1500 * Create a new input stream structure.
1501 *
1502 * Returns the new input stream or NULL
1503 */
1504 xmlParserInputPtr
1505 xmlNewInputStream(xmlParserCtxtPtr ctxt) {
1506 xmlParserInputPtr input;
1507
1508 input = (xmlParserInputPtr) xmlMalloc(sizeof(xmlParserInput));
1509 if (input == NULL) {
1510 xmlErrMemory(ctxt, "couldn't allocate a new input stream\n");
1511 return(NULL);
1512 }
1513 memset(input, 0, sizeof(xmlParserInput));
1514 input->line = 1;
1515 input->col = 1;
1516
1517 /*
1518 * If the context is NULL the id cannot be initialized, but that
1519 * should not happen while parsing which is the situation where
1520 * the id is actually needed.
1521 */
1522 if (ctxt != NULL) {
1523 if (input->id >= INT_MAX) {
1524 xmlErrMemory(ctxt, "Input ID overflow\n");
1525 return(NULL);
1526 }
1527 input->id = ctxt->input_id++;
1528 }
1529
1530 return(input);
1531 }
1532
1533 /**
1534 * xmlNewIOInputStream:
1535 * @ctxt: an XML parser context
1536 * @input: an I/O Input
1537 * @enc: the charset encoding if known
1538 *
1539 * Create a new input stream structure encapsulating the @input into
1540 * a stream suitable for the parser.
1541 *
1542 * Returns the new input stream or NULL
1543 */
1544 xmlParserInputPtr
1545 xmlNewIOInputStream(xmlParserCtxtPtr ctxt, xmlParserInputBufferPtr input,
1546 xmlCharEncoding enc) {
1547 xmlParserInputPtr inputStream;
1548
1549 if (input == NULL) return(NULL);
1550 if (xmlParserDebugEntities)
1551 xmlGenericError(xmlGenericErrorContext, "new input from I/O\n");
1552 inputStream = xmlNewInputStream(ctxt);
1553 if (inputStream == NULL) {
1554 return(NULL);
1555 }
1556 inputStream->filename = NULL;
1557 inputStream->buf = input;
1558 xmlBufResetInput(inputStream->buf->buffer, inputStream);
1559
1560 if (enc != XML_CHAR_ENCODING_NONE) {
1561 xmlSwitchEncoding(ctxt, enc);
1562 }
1563
1564 return(inputStream);
1565 }
1566
1567 /**
1568 * xmlNewEntityInputStream:
1569 * @ctxt: an XML parser context
1570 * @entity: an Entity pointer
1571 *
1572 * DEPRECATED: Internal function, do not use.
1573 *
1574 * Create a new input stream based on an xmlEntityPtr
1575 *
1576 * Returns the new input stream or NULL
1577 */
1578 xmlParserInputPtr
1579 xmlNewEntityInputStream(xmlParserCtxtPtr ctxt, xmlEntityPtr entity) {
1580 xmlParserInputPtr input;
1581
1582 if (entity == NULL) {
1583 xmlErrInternal(ctxt, "xmlNewEntityInputStream entity = NULL\n",
1584 NULL);
1585 return(NULL);
1586 }
1587 if (xmlParserDebugEntities)
1588 xmlGenericError(xmlGenericErrorContext,
1589 "new input from entity: %s\n", entity->name);
1590 if (entity->content == NULL) {
1591 switch (entity->etype) {
1592 case XML_EXTERNAL_GENERAL_UNPARSED_ENTITY:
1593 xmlErrInternal(ctxt, "Cannot parse entity %s\n",
1594 entity->name);
1595 break;
1596 case XML_EXTERNAL_GENERAL_PARSED_ENTITY:
1597 case XML_EXTERNAL_PARAMETER_ENTITY:
1598 input = xmlLoadExternalEntity((char *) entity->URI,
1599 (char *) entity->ExternalID, ctxt);
1600 if (input != NULL)
1601 input->entity = entity;
1602 return(input);
1603 case XML_INTERNAL_GENERAL_ENTITY:
1604 xmlErrInternal(ctxt,
1605 "Internal entity %s without content !\n",
1606 entity->name);
1607 break;
1608 case XML_INTERNAL_PARAMETER_ENTITY:
1609 xmlErrInternal(ctxt,
1610 "Internal parameter entity %s without content !\n",
1611 entity->name);
1612 break;
1613 case XML_INTERNAL_PREDEFINED_ENTITY:
1614 xmlErrInternal(ctxt,
1615 "Predefined entity %s without content !\n",
1616 entity->name);
1617 break;
1618 }
1619 return(NULL);
1620 }
1621 input = xmlNewInputStream(ctxt);
1622 if (input == NULL) {
1623 return(NULL);
1624 }
1625 if (entity->URI != NULL)
1626 input->filename = (char *) xmlStrdup((xmlChar *) entity->URI);
1627 input->base = entity->content;
1628 if (entity->length == 0)
1629 entity->length = xmlStrlen(entity->content);
1630 input->cur = entity->content;
1631 input->length = entity->length;
1632 input->end = &entity->content[input->length];
1633 input->entity = entity;
1634 return(input);
1635 }
1636
1637 /**
1638 * xmlNewStringInputStream:
1639 * @ctxt: an XML parser context
1640 * @buffer: an memory buffer
1641 *
1642 * Create a new input stream based on a memory buffer.
1643 * Returns the new input stream
1644 */
1645 xmlParserInputPtr
1646 xmlNewStringInputStream(xmlParserCtxtPtr ctxt, const xmlChar *buffer) {
1647 xmlParserInputPtr input;
1648 xmlParserInputBufferPtr buf;
1649
1650 if (buffer == NULL) {
1651 xmlErrInternal(ctxt, "xmlNewStringInputStream string = NULL\n",
1652 NULL);
1653 return(NULL);
1654 }
1655 if (xmlParserDebugEntities)
1656 xmlGenericError(xmlGenericErrorContext,
1657 "new fixed input: %.30s\n", buffer);
1658 buf = xmlParserInputBufferCreateString(buffer);
1659 if (buf == NULL) {
1660 xmlErrMemory(ctxt, NULL);
1661 return(NULL);
1662 }
1663 input = xmlNewInputStream(ctxt);
1664 if (input == NULL) {
1665 xmlErrMemory(ctxt, "couldn't allocate a new input stream\n");
1666 xmlFreeParserInputBuffer(buf);
1667 return(NULL);
1668 }
1669 input->buf = buf;
1670 xmlBufResetInput(input->buf->buffer, input);
1671 return(input);
1672 }
1673
1674 /**
1675 * xmlNewInputFromFile:
1676 * @ctxt: an XML parser context
1677 * @filename: the filename to use as entity
1678 *
1679 * Create a new input stream based on a file or an URL.
1680 *
1681 * Returns the new input stream or NULL in case of error
1682 */
1683 xmlParserInputPtr
1684 xmlNewInputFromFile(xmlParserCtxtPtr ctxt, const char *filename) {
1685 xmlParserInputBufferPtr buf;
1686 xmlParserInputPtr inputStream;
1687 char *directory = NULL;
1688 xmlChar *URI = NULL;
1689
1690 if (xmlParserDebugEntities)
1691 xmlGenericError(xmlGenericErrorContext,
1692 "new input from file: %s\n", filename);
1693 if (ctxt == NULL) return(NULL);
1694 buf = xmlParserInputBufferCreateFilename(filename, XML_CHAR_ENCODING_NONE);
1695 if (buf == NULL) {
1696 if (filename == NULL)
1697 __xmlLoaderErr(ctxt,
1698 "failed to load external entity: NULL filename \n",
1699 NULL);
1700 else
1701 __xmlLoaderErr(ctxt, "failed to load external entity \"%s\"\n",
1702 (const char *) filename);
1703 return(NULL);
1704 }
1705
1706 inputStream = xmlNewInputStream(ctxt);
1707 if (inputStream == NULL) {
1708 xmlFreeParserInputBuffer(buf);
1709 return(NULL);
1710 }
1711
1712 inputStream->buf = buf;
1713 inputStream = xmlCheckHTTPInput(ctxt, inputStream);
1714 if (inputStream == NULL)
1715 return(NULL);
1716
1717 if (inputStream->filename == NULL)
1718 URI = xmlStrdup((xmlChar *) filename);
1719 else
1720 URI = xmlStrdup((xmlChar *) inputStream->filename);
1721 directory = xmlParserGetDirectory((const char *) URI);
1722 if (inputStream->filename != NULL) xmlFree((char *)inputStream->filename);
1723 inputStream->filename = (char *) xmlCanonicPath((const xmlChar *) URI);
1724 if (URI != NULL) xmlFree((char *) URI);
1725 inputStream->directory = directory;
1726
1727 xmlBufResetInput(inputStream->buf->buffer, inputStream);
1728 if ((ctxt->directory == NULL) && (directory != NULL))
1729 ctxt->directory = (char *) xmlStrdup((const xmlChar *) directory);
1730 return(inputStream);
1731 }
1732
1733 /************************************************************************
1734 * *
1735 * Commodity functions to handle parser contexts *
1736 * *
1737 ************************************************************************/
1738
1739 /**
1740 * xmlInitSAXParserCtxt:
1741 * @ctxt: XML parser context
1742 * @sax: SAX handlert
1743 * @userData: user data
1744 *
1745 * Initialize a SAX parser context
1746 *
1747 * Returns 0 in case of success and -1 in case of error
1748 */
1749
1750 static int
1751 xmlInitSAXParserCtxt(xmlParserCtxtPtr ctxt, const xmlSAXHandler *sax,
1752 void *userData)
1753 {
1754 xmlParserInputPtr input;
1755
1756 if(ctxt==NULL) {
1757 xmlErrInternal(NULL, "Got NULL parser context\n", NULL);
1758 return(-1);
1759 }
1760
1761 xmlInitParser();
1762
1763 if (ctxt->dict == NULL)
1764 ctxt->dict = xmlDictCreate();
1765 if (ctxt->dict == NULL) {
1766 xmlErrMemory(NULL, "cannot initialize parser context\n");
1767 return(-1);
1768 }
1769 xmlDictSetLimit(ctxt->dict, XML_MAX_DICTIONARY_LIMIT);
1770
1771 if (ctxt->sax == NULL)
1772 ctxt->sax = (xmlSAXHandler *) xmlMalloc(sizeof(xmlSAXHandler));
1773 if (ctxt->sax == NULL) {
1774 xmlErrMemory(NULL, "cannot initialize parser context\n");
1775 return(-1);
1776 }
1777 if (sax == NULL) {
1778 memset(ctxt->sax, 0, sizeof(xmlSAXHandler));
1779 xmlSAXVersion(ctxt->sax, 2);
1780 ctxt->userData = ctxt;
1781 } else {
1782 if (sax->initialized == XML_SAX2_MAGIC) {
1783 memcpy(ctxt->sax, sax, sizeof(xmlSAXHandler));
1784 } else {
1785 memset(ctxt->sax, 0, sizeof(xmlSAXHandler));
1786 memcpy(ctxt->sax, sax, sizeof(xmlSAXHandlerV1));
1787 }
1788 ctxt->userData = userData ? userData : ctxt;
1789 }
1790
1791 ctxt->maxatts = 0;
1792 ctxt->atts = NULL;
1793 /* Allocate the Input stack */
1794 if (ctxt->inputTab == NULL) {
1795 ctxt->inputTab = (xmlParserInputPtr *)
1796 xmlMalloc(5 * sizeof(xmlParserInputPtr));
1797 ctxt->inputMax = 5;
1798 }
1799 if (ctxt->inputTab == NULL) {
1800 xmlErrMemory(NULL, "cannot initialize parser context\n");
1801 ctxt->inputNr = 0;
1802 ctxt->inputMax = 0;
1803 ctxt->input = NULL;
1804 return(-1);
1805 }
1806 while ((input = inputPop(ctxt)) != NULL) { /* Non consuming */
1807 xmlFreeInputStream(input);
1808 }
1809 ctxt->inputNr = 0;
1810 ctxt->input = NULL;
1811
1812 ctxt->version = NULL;
1813 ctxt->encoding = NULL;
1814 ctxt->standalone = -1;
1815 ctxt->hasExternalSubset = 0;
1816 ctxt->hasPErefs = 0;
1817 ctxt->html = 0;
1818 ctxt->external = 0;
1819 ctxt->instate = XML_PARSER_START;
1820 ctxt->token = 0;
1821 ctxt->directory = NULL;
1822
1823 /* Allocate the Node stack */
1824 if (ctxt->nodeTab == NULL) {
1825 ctxt->nodeTab = (xmlNodePtr *) xmlMalloc(10 * sizeof(xmlNodePtr));
1826 ctxt->nodeMax = 10;
1827 }
1828 if (ctxt->nodeTab == NULL) {
1829 xmlErrMemory(NULL, "cannot initialize parser context\n");
1830 ctxt->nodeNr = 0;
1831 ctxt->nodeMax = 0;
1832 ctxt->node = NULL;
1833 ctxt->inputNr = 0;
1834 ctxt->inputMax = 0;
1835 ctxt->input = NULL;
1836 return(-1);
1837 }
1838 ctxt->nodeNr = 0;
1839 ctxt->node = NULL;
1840
1841 /* Allocate the Name stack */
1842 if (ctxt->nameTab == NULL) {
1843 ctxt->nameTab = (const xmlChar **) xmlMalloc(10 * sizeof(xmlChar *));
1844 ctxt->nameMax = 10;
1845 }
1846 if (ctxt->nameTab == NULL) {
1847 xmlErrMemory(NULL, "cannot initialize parser context\n");
1848 ctxt->nodeNr = 0;
1849 ctxt->nodeMax = 0;
1850 ctxt->node = NULL;
1851 ctxt->inputNr = 0;
1852 ctxt->inputMax = 0;
1853 ctxt->input = NULL;
1854 ctxt->nameNr = 0;
1855 ctxt->nameMax = 0;
1856 ctxt->name = NULL;
1857 return(-1);
1858 }
1859 ctxt->nameNr = 0;
1860 ctxt->name = NULL;
1861
1862 /* Allocate the space stack */
1863 if (ctxt->spaceTab == NULL) {
1864 ctxt->spaceTab = (int *) xmlMalloc(10 * sizeof(int));
1865 ctxt->spaceMax = 10;
1866 }
1867 if (ctxt->spaceTab == NULL) {
1868 xmlErrMemory(NULL, "cannot initialize parser context\n");
1869 ctxt->nodeNr = 0;
1870 ctxt->nodeMax = 0;
1871 ctxt->node = NULL;
1872 ctxt->inputNr = 0;
1873 ctxt->inputMax = 0;
1874 ctxt->input = NULL;
1875 ctxt->nameNr = 0;
1876 ctxt->nameMax = 0;
1877 ctxt->name = NULL;
1878 ctxt->spaceNr = 0;
1879 ctxt->spaceMax = 0;
1880 ctxt->space = NULL;
1881 return(-1);
1882 }
1883 ctxt->spaceNr = 1;
1884 ctxt->spaceMax = 10;
1885 ctxt->spaceTab[0] = -1;
1886 ctxt->space = &ctxt->spaceTab[0];
1887 ctxt->myDoc = NULL;
1888 ctxt->wellFormed = 1;
1889 ctxt->nsWellFormed = 1;
1890 ctxt->valid = 1;
1891 ctxt->loadsubset = xmlLoadExtDtdDefaultValue;
1892 if (ctxt->loadsubset) {
1893 ctxt->options |= XML_PARSE_DTDLOAD;
1894 }
1895 ctxt->validate = xmlDoValidityCheckingDefaultValue;
1896 ctxt->pedantic = xmlPedanticParserDefaultValue;
1897 if (ctxt->pedantic) {
1898 ctxt->options |= XML_PARSE_PEDANTIC;
1899 }
1900 ctxt->linenumbers = xmlLineNumbersDefaultValue;
1901 ctxt->keepBlanks = xmlKeepBlanksDefaultValue;
1902 if (ctxt->keepBlanks == 0) {
1903 ctxt->sax->ignorableWhitespace = xmlSAX2IgnorableWhitespace;
1904 ctxt->options |= XML_PARSE_NOBLANKS;
1905 }
1906
1907 ctxt->vctxt.flags = XML_VCTXT_USE_PCTXT;
1908 ctxt->vctxt.userData = ctxt;
1909 ctxt->vctxt.error = xmlParserValidityError;
1910 ctxt->vctxt.warning = xmlParserValidityWarning;
1911 if (ctxt->validate) {
1912 if (xmlGetWarningsDefaultValue == 0)
1913 ctxt->vctxt.warning = NULL;
1914 else
1915 ctxt->vctxt.warning = xmlParserValidityWarning;
1916 ctxt->vctxt.nodeMax = 0;
1917 ctxt->options |= XML_PARSE_DTDVALID;
1918 }
1919 ctxt->replaceEntities = xmlSubstituteEntitiesDefaultValue;
1920 if (ctxt->replaceEntities) {
1921 ctxt->options |= XML_PARSE_NOENT;
1922 }
1923 ctxt->record_info = 0;
1924 ctxt->checkIndex = 0;
1925 ctxt->inSubset = 0;
1926 ctxt->errNo = XML_ERR_OK;
1927 ctxt->depth = 0;
1928 ctxt->catalogs = NULL;
1929 ctxt->sizeentities = 0;
1930 ctxt->sizeentcopy = 0;
1931 ctxt->input_id = 1;
1932 ctxt->maxAmpl = XML_MAX_AMPLIFICATION_DEFAULT;
1933 xmlInitNodeInfoSeq(&ctxt->node_seq);
1934
1935 if (ctxt->nsdb == NULL) {
1936 ctxt->nsdb = xmlParserNsCreate();
1937 if (ctxt->nsdb == NULL) {
1938 xmlErrMemory(ctxt, NULL);
1939 return(-1);
1940 }
1941 }
1942
1943 return(0);
1944 }
1945
1946 /**
1947 * xmlInitParserCtxt:
1948 * @ctxt: an XML parser context
1949 *
1950 * DEPRECATED: Internal function which will be made private in a future
1951 * version.
1952 *
1953 * Initialize a parser context
1954 *
1955 * Returns 0 in case of success and -1 in case of error
1956 */
1957
1958 int
1959 xmlInitParserCtxt(xmlParserCtxtPtr ctxt)
1960 {
1961 return(xmlInitSAXParserCtxt(ctxt, NULL, NULL));
1962 }
1963
1964 /**
1965 * xmlFreeParserCtxt:
1966 * @ctxt: an XML parser context
1967 *
1968 * Free all the memory used by a parser context. However the parsed
1969 * document in ctxt->myDoc is not freed.
1970 */
1971
1972 void
1973 xmlFreeParserCtxt(xmlParserCtxtPtr ctxt)
1974 {
1975 xmlParserInputPtr input;
1976
1977 if (ctxt == NULL) return;
1978
1979 while ((input = inputPop(ctxt)) != NULL) { /* Non consuming */
1980 xmlFreeInputStream(input);
1981 }
1982 if (ctxt->spaceTab != NULL) xmlFree(ctxt->spaceTab);
1983 if (ctxt->nameTab != NULL) xmlFree((xmlChar * *)ctxt->nameTab);
1984 if (ctxt->nodeTab != NULL) xmlFree(ctxt->nodeTab);
1985 if (ctxt->nodeInfoTab != NULL) xmlFree(ctxt->nodeInfoTab);
1986 if (ctxt->inputTab != NULL) xmlFree(ctxt->inputTab);
1987 if (ctxt->version != NULL) xmlFree((char *) ctxt->version);
1988 if (ctxt->encoding != NULL) xmlFree((char *) ctxt->encoding);
1989 if (ctxt->extSubURI != NULL) xmlFree((char *) ctxt->extSubURI);
1990 if (ctxt->extSubSystem != NULL) xmlFree((char *) ctxt->extSubSystem);
1991 #ifdef LIBXML_SAX1_ENABLED
1992 if ((ctxt->sax != NULL) &&
1993 (ctxt->sax != (xmlSAXHandlerPtr) &xmlDefaultSAXHandler))
1994 #else
1995 if (ctxt->sax != NULL)
1996 #endif /* LIBXML_SAX1_ENABLED */
1997 xmlFree(ctxt->sax);
1998 if (ctxt->directory != NULL) xmlFree((char *) ctxt->directory);
1999 if (ctxt->vctxt.nodeTab != NULL) xmlFree(ctxt->vctxt.nodeTab);
2000 if (ctxt->atts != NULL) xmlFree((xmlChar * *)ctxt->atts);
2001 if (ctxt->dict != NULL) xmlDictFree(ctxt->dict);
2002 if (ctxt->nsTab != NULL) xmlFree(ctxt->nsTab);
2003 if (ctxt->nsdb != NULL) xmlParserNsFree(ctxt->nsdb);
2004 if (ctxt->attrHash != NULL) xmlFree(ctxt->attrHash);
2005 if (ctxt->pushTab != NULL) xmlFree(ctxt->pushTab);
2006 if (ctxt->attallocs != NULL) xmlFree(ctxt->attallocs);
2007 if (ctxt->attsDefault != NULL)
2008 xmlHashFree(ctxt->attsDefault, xmlHashDefaultDeallocator);
2009 if (ctxt->attsSpecial != NULL)
2010 xmlHashFree(ctxt->attsSpecial, NULL);
2011 if (ctxt->freeElems != NULL) {
2012 xmlNodePtr cur, next;
2013
2014 cur = ctxt->freeElems;
2015 while (cur != NULL) {
2016 next = cur->next;
2017 xmlFree(cur);
2018 cur = next;
2019 }
2020 }
2021 if (ctxt->freeAttrs != NULL) {
2022 xmlAttrPtr cur, next;
2023
2024 cur = ctxt->freeAttrs;
2025 while (cur != NULL) {
2026 next = cur->next;
2027 xmlFree(cur);
2028 cur = next;
2029 }
2030 }
2031 /*
2032 * cleanup the error strings
2033 */
2034 if (ctxt->lastError.message != NULL)
2035 xmlFree(ctxt->lastError.message);
2036 if (ctxt->lastError.file != NULL)
2037 xmlFree(ctxt->lastError.file);
2038 if (ctxt->lastError.str1 != NULL)
2039 xmlFree(ctxt->lastError.str1);
2040 if (ctxt->lastError.str2 != NULL)
2041 xmlFree(ctxt->lastError.str2);
2042 if (ctxt->lastError.str3 != NULL)
2043 xmlFree(ctxt->lastError.str3);
2044
2045 #ifdef LIBXML_CATALOG_ENABLED
2046 if (ctxt->catalogs != NULL)
2047 xmlCatalogFreeLocal(ctxt->catalogs);
2048 #endif
2049 xmlFree(ctxt);
2050 }
2051
2052 /**
2053 * xmlNewParserCtxt:
2054 *
2055 * Allocate and initialize a new parser context.
2056 *
2057 * Returns the xmlParserCtxtPtr or NULL
2058 */
2059
2060 xmlParserCtxtPtr
2061 xmlNewParserCtxt(void)
2062 {
2063 return(xmlNewSAXParserCtxt(NULL, NULL));
2064 }
2065
2066 /**
2067 * xmlNewSAXParserCtxt:
2068 * @sax: SAX handler
2069 * @userData: user data
2070 *
2071 * Allocate and initialize a new SAX parser context. If userData is NULL,
2072 * the parser context will be passed as user data.
2073 *
2074 * Returns the xmlParserCtxtPtr or NULL if memory allocation failed.
2075 */
2076
2077 xmlParserCtxtPtr
2078 xmlNewSAXParserCtxt(const xmlSAXHandler *sax, void *userData)
2079 {
2080 xmlParserCtxtPtr ctxt;
2081
2082 ctxt = (xmlParserCtxtPtr) xmlMalloc(sizeof(xmlParserCtxt));
2083 if (ctxt == NULL) {
2084 xmlErrMemory(NULL, "cannot allocate parser context\n");
2085 return(NULL);
2086 }
2087 memset(ctxt, 0, sizeof(xmlParserCtxt));
2088 if (xmlInitSAXParserCtxt(ctxt, sax, userData) < 0) {
2089 xmlFreeParserCtxt(ctxt);
2090 return(NULL);
2091 }
2092 return(ctxt);
2093 }
2094
2095 /************************************************************************
2096 * *
2097 * Handling of node information *
2098 * *
2099 ************************************************************************/
2100
2101 /**
2102 * xmlClearParserCtxt:
2103 * @ctxt: an XML parser context
2104 *
2105 * Clear (release owned resources) and reinitialize a parser context
2106 */
2107
2108 void
2109 xmlClearParserCtxt(xmlParserCtxtPtr ctxt)
2110 {
2111 if (ctxt==NULL)
2112 return;
2113 xmlClearNodeInfoSeq(&ctxt->node_seq);
2114 xmlCtxtReset(ctxt);
2115 }
2116
2117
2118 /**
2119 * xmlParserFindNodeInfo:
2120 * @ctx: an XML parser context
2121 * @node: an XML node within the tree
2122 *
2123 * DEPRECATED: Don't use.
2124 *
2125 * Find the parser node info struct for a given node
2126 *
2127 * Returns an xmlParserNodeInfo block pointer or NULL
2128 */
2129 const xmlParserNodeInfo *
2130 xmlParserFindNodeInfo(const xmlParserCtxtPtr ctx, const xmlNodePtr node)
2131 {
2132 unsigned long pos;
2133
2134 if ((ctx == NULL) || (node == NULL))
2135 return (NULL);
2136 /* Find position where node should be at */
2137 pos = xmlParserFindNodeInfoIndex(&ctx->node_seq, node);
2138 if (pos < ctx->node_seq.length
2139 && ctx->node_seq.buffer[pos].node == node)
2140 return &ctx->node_seq.buffer[pos];
2141 else
2142 return NULL;
2143 }
2144
2145
2146 /**
2147 * xmlInitNodeInfoSeq:
2148 * @seq: a node info sequence pointer
2149 *
2150 * DEPRECATED: Don't use.
2151 *
2152 * -- Initialize (set to initial state) node info sequence
2153 */
2154 void
2155 xmlInitNodeInfoSeq(xmlParserNodeInfoSeqPtr seq)
2156 {
2157 if (seq == NULL)
2158 return;
2159 seq->length = 0;
2160 seq->maximum = 0;
2161 seq->buffer = NULL;
2162 }
2163
2164 /**
2165 * xmlClearNodeInfoSeq:
2166 * @seq: a node info sequence pointer
2167 *
2168 * DEPRECATED: Don't use.
2169 *
2170 * -- Clear (release memory and reinitialize) node
2171 * info sequence
2172 */
2173 void
2174 xmlClearNodeInfoSeq(xmlParserNodeInfoSeqPtr seq)
2175 {
2176 if (seq == NULL)
2177 return;
2178 if (seq->buffer != NULL)
2179 xmlFree(seq->buffer);
2180 xmlInitNodeInfoSeq(seq);
2181 }
2182
2183 /**
2184 * xmlParserFindNodeInfoIndex:
2185 * @seq: a node info sequence pointer
2186 * @node: an XML node pointer
2187 *
2188 * DEPRECATED: Don't use.
2189 *
2190 * xmlParserFindNodeInfoIndex : Find the index that the info record for
2191 * the given node is or should be at in a sorted sequence
2192 *
2193 * Returns a long indicating the position of the record
2194 */
2195 unsigned long
2196 xmlParserFindNodeInfoIndex(const xmlParserNodeInfoSeqPtr seq,
2197 const xmlNodePtr node)
2198 {
2199 unsigned long upper, lower, middle;
2200 int found = 0;
2201
2202 if ((seq == NULL) || (node == NULL))
2203 return ((unsigned long) -1);
2204
2205 /* Do a binary search for the key */
2206 lower = 1;
2207 upper = seq->length;
2208 middle = 0;
2209 while (lower <= upper && !found) {
2210 middle = lower + (upper - lower) / 2;
2211 if (node == seq->buffer[middle - 1].node)
2212 found = 1;
2213 else if (node < seq->buffer[middle - 1].node)
2214 upper = middle - 1;
2215 else
2216 lower = middle + 1;
2217 }
2218
2219 /* Return position */
2220 if (middle == 0 || seq->buffer[middle - 1].node < node)
2221 return middle;
2222 else
2223 return middle - 1;
2224 }
2225
2226
2227 /**
2228 * xmlParserAddNodeInfo:
2229 * @ctxt: an XML parser context
2230 * @info: a node info sequence pointer
2231 *
2232 * DEPRECATED: Don't use.
2233 *
2234 * Insert node info record into the sorted sequence
2235 */
2236 void
2237 xmlParserAddNodeInfo(xmlParserCtxtPtr ctxt,
2238 const xmlParserNodeInfoPtr info)
2239 {
2240 unsigned long pos;
2241
2242 if ((ctxt == NULL) || (info == NULL)) return;
2243
2244 /* Find pos and check to see if node is already in the sequence */
2245 pos = xmlParserFindNodeInfoIndex(&ctxt->node_seq, (xmlNodePtr)
2246 info->node);
2247
2248 if ((pos < ctxt->node_seq.length) &&
2249 (ctxt->node_seq.buffer != NULL) &&
2250 (ctxt->node_seq.buffer[pos].node == info->node)) {
2251 ctxt->node_seq.buffer[pos] = *info;
2252 }
2253
2254 /* Otherwise, we need to add new node to buffer */
2255 else {
2256 if ((ctxt->node_seq.length + 1 > ctxt->node_seq.maximum) ||
2257 (ctxt->node_seq.buffer == NULL)) {
2258 xmlParserNodeInfo *tmp_buffer;
2259 unsigned int byte_size;
2260
2261 if (ctxt->node_seq.maximum == 0)
2262 ctxt->node_seq.maximum = 2;
2263 byte_size = (sizeof(*ctxt->node_seq.buffer) *
2264 (2 * ctxt->node_seq.maximum));
2265
2266 if (ctxt->node_seq.buffer == NULL)
2267 tmp_buffer = (xmlParserNodeInfo *) xmlMalloc(byte_size);
2268 else
2269 tmp_buffer =
2270 (xmlParserNodeInfo *) xmlRealloc(ctxt->node_seq.buffer,
2271 byte_size);
2272
2273 if (tmp_buffer == NULL) {
2274 xmlErrMemory(ctxt, "failed to allocate buffer\n");
2275 return;
2276 }
2277 ctxt->node_seq.buffer = tmp_buffer;
2278 ctxt->node_seq.maximum *= 2;
2279 }
2280
2281 /* If position is not at end, move elements out of the way */
2282 if (pos != ctxt->node_seq.length) {
2283 unsigned long i;
2284
2285 for (i = ctxt->node_seq.length; i > pos; i--)
2286 ctxt->node_seq.buffer[i] = ctxt->node_seq.buffer[i - 1];
2287 }
2288
2289 /* Copy element and increase length */
2290 ctxt->node_seq.buffer[pos] = *info;
2291 ctxt->node_seq.length++;
2292 }
2293 }
2294
2295 /************************************************************************
2296 * *
2297 * Defaults settings *
2298 * *
2299 ************************************************************************/
2300 /**
2301 * xmlPedanticParserDefault:
2302 * @val: int 0 or 1
2303 *
2304 * DEPRECATED: Use the modern options API with XML_PARSE_PEDANTIC.
2305 *
2306 * Set and return the previous value for enabling pedantic warnings.
2307 *
2308 * Returns the last value for 0 for no substitution, 1 for substitution.
2309 */
2310
2311 int
2312 xmlPedanticParserDefault(int val) {
2313 int old = xmlPedanticParserDefaultValue;
2314
2315 xmlPedanticParserDefaultValue = val;
2316 return(old);
2317 }
2318
2319 /**
2320 * xmlLineNumbersDefault:
2321 * @val: int 0 or 1
2322 *
2323 * DEPRECATED: The modern options API always enables line numbers.
2324 *
2325 * Set and return the previous value for enabling line numbers in elements
2326 * contents. This may break on old application and is turned off by default.
2327 *
2328 * Returns the last value for 0 for no substitution, 1 for substitution.
2329 */
2330
2331 int
2332 xmlLineNumbersDefault(int val) {
2333 int old = xmlLineNumbersDefaultValue;
2334
2335 xmlLineNumbersDefaultValue = val;
2336 return(old);
2337 }
2338
2339 /**
2340 * xmlSubstituteEntitiesDefault:
2341 * @val: int 0 or 1
2342 *
2343 * DEPRECATED: Use the modern options API with XML_PARSE_NOENT.
2344 *
2345 * Set and return the previous value for default entity support.
2346 * Initially the parser always keep entity references instead of substituting
2347 * entity values in the output. This function has to be used to change the
2348 * default parser behavior
2349 * SAX::substituteEntities() has to be used for changing that on a file by
2350 * file basis.
2351 *
2352 * Returns the last value for 0 for no substitution, 1 for substitution.
2353 */
2354
2355 int
2356 xmlSubstituteEntitiesDefault(int val) {
2357 int old = xmlSubstituteEntitiesDefaultValue;
2358
2359 xmlSubstituteEntitiesDefaultValue = val;
2360 return(old);
2361 }
2362
2363 /**
2364 * xmlKeepBlanksDefault:
2365 * @val: int 0 or 1
2366 *
2367 * DEPRECATED: Use the modern options API with XML_PARSE_NOBLANKS.
2368 *
2369 * Set and return the previous value for default blanks text nodes support.
2370 * The 1.x version of the parser used an heuristic to try to detect
2371 * ignorable white spaces. As a result the SAX callback was generating
2372 * xmlSAX2IgnorableWhitespace() callbacks instead of characters() one, and when
2373 * using the DOM output text nodes containing those blanks were not generated.
2374 * The 2.x and later version will switch to the XML standard way and
2375 * ignorableWhitespace() are only generated when running the parser in
2376 * validating mode and when the current element doesn't allow CDATA or
2377 * mixed content.
2378 * This function is provided as a way to force the standard behavior
2379 * on 1.X libs and to switch back to the old mode for compatibility when
2380 * running 1.X client code on 2.X . Upgrade of 1.X code should be done
2381 * by using xmlIsBlankNode() commodity function to detect the "empty"
2382 * nodes generated.
2383 * This value also affect autogeneration of indentation when saving code
2384 * if blanks sections are kept, indentation is not generated.
2385 *
2386 * Returns the last value for 0 for no substitution, 1 for substitution.
2387 */
2388
2389 int
2390 xmlKeepBlanksDefault(int val) {
2391 int old = xmlKeepBlanksDefaultValue;
2392
2393 xmlKeepBlanksDefaultValue = val;
2394 #ifdef LIBXML_OUTPUT_ENABLED
2395 if (!val)
2396 xmlIndentTreeOutput = 1;
2397 #endif
2398 return(old);
2399 }
2400