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) && !defined (__CYGWIN__)
14 #define XML_DIR_SEP '\\'
15 #else
16 #define XML_DIR_SEP '/'
17 #endif
18
19 #include <string.h>
20 #ifdef HAVE_CTYPE_H
21 #include <ctype.h>
22 #endif
23 #ifdef HAVE_STDLIB_H
24 #include <stdlib.h>
25 #endif
26 #ifdef HAVE_SYS_STAT_H
27 #include <sys/stat.h>
28 #endif
29 #ifdef HAVE_FCNTL_H
30 #include <fcntl.h>
31 #endif
32 #ifdef HAVE_UNISTD_H
33 #include <unistd.h>
34 #endif
35 #ifdef HAVE_ZLIB_H
36 #include <zlib.h>
37 #endif
38
39 #include <libxml/xmlmemory.h>
40 #include <libxml/tree.h>
41 #include <libxml/parser.h>
42 #include <libxml/parserInternals.h>
43 #include <libxml/valid.h>
44 #include <libxml/entities.h>
45 #include <libxml/xmlerror.h>
46 #include <libxml/encoding.h>
47 #include <libxml/valid.h>
48 #include <libxml/xmlIO.h>
49 #include <libxml/uri.h>
50 #include <libxml/dict.h>
51 #include <libxml/SAX.h>
52 #ifdef LIBXML_CATALOG_ENABLED
53 #include <libxml/catalog.h>
54 #endif
55 #include <libxml/globals.h>
56 #include <libxml/chvalid.h>
57
58 #include "buf.h"
59 #include "enc.h"
60
61 /*
62 * Various global defaults for parsing
63 */
64
65 /**
66 * xmlCheckVersion:
67 * @version: the include version number
68 *
69 * check the compiled lib version against the include one.
70 * This can warn or immediately kill the application
71 */
72 void
xmlCheckVersion(int version)73 xmlCheckVersion(int version) {
74 int myversion = (int) LIBXML_VERSION;
75
76 xmlInitParser();
77
78 if ((myversion / 10000) != (version / 10000)) {
79 xmlGenericError(xmlGenericErrorContext,
80 "Fatal: program compiled against libxml %d using libxml %d\n",
81 (version / 10000), (myversion / 10000));
82 fprintf(stderr,
83 "Fatal: program compiled against libxml %d using libxml %d\n",
84 (version / 10000), (myversion / 10000));
85 }
86 if ((myversion / 100) < (version / 100)) {
87 xmlGenericError(xmlGenericErrorContext,
88 "Warning: program compiled against libxml %d using older %d\n",
89 (version / 100), (myversion / 100));
90 }
91 }
92
93
94 /************************************************************************
95 * *
96 * Some factorized error routines *
97 * *
98 ************************************************************************/
99
100
101 /**
102 * xmlErrMemory:
103 * @ctxt: an XML parser context
104 * @extra: extra informations
105 *
106 * Handle a redefinition of attribute error
107 */
108 void
xmlErrMemory(xmlParserCtxtPtr ctxt,const char * extra)109 xmlErrMemory(xmlParserCtxtPtr ctxt, const char *extra)
110 {
111 if ((ctxt != NULL) && (ctxt->disableSAX != 0) &&
112 (ctxt->instate == XML_PARSER_EOF))
113 return;
114 if (ctxt != NULL) {
115 ctxt->errNo = XML_ERR_NO_MEMORY;
116 ctxt->instate = XML_PARSER_EOF;
117 ctxt->disableSAX = 1;
118 }
119 if (extra)
120 __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER,
121 XML_ERR_NO_MEMORY, XML_ERR_FATAL, NULL, 0, extra,
122 NULL, NULL, 0, 0,
123 "Memory allocation failed : %s\n", extra);
124 else
125 __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER,
126 XML_ERR_NO_MEMORY, XML_ERR_FATAL, NULL, 0, NULL,
127 NULL, NULL, 0, 0, "Memory allocation failed\n");
128 }
129
130 /**
131 * __xmlErrEncoding:
132 * @ctxt: an XML parser context
133 * @xmlerr: the error number
134 * @msg: the error message
135 * @str1: an string info
136 * @str2: an string info
137 *
138 * Handle an encoding error
139 */
140 void
__xmlErrEncoding(xmlParserCtxtPtr ctxt,xmlParserErrors xmlerr,const char * msg,const xmlChar * str1,const xmlChar * str2)141 __xmlErrEncoding(xmlParserCtxtPtr ctxt, xmlParserErrors xmlerr,
142 const char *msg, const xmlChar * str1, const xmlChar * str2)
143 {
144 if ((ctxt != NULL) && (ctxt->disableSAX != 0) &&
145 (ctxt->instate == XML_PARSER_EOF))
146 return;
147 if (ctxt != NULL)
148 ctxt->errNo = xmlerr;
149 __xmlRaiseError(NULL, NULL, NULL,
150 ctxt, NULL, XML_FROM_PARSER, xmlerr, XML_ERR_FATAL,
151 NULL, 0, (const char *) str1, (const char *) str2,
152 NULL, 0, 0, msg, str1, str2);
153 if (ctxt != NULL) {
154 ctxt->wellFormed = 0;
155 if (ctxt->recovery == 0)
156 ctxt->disableSAX = 1;
157 }
158 }
159
160 /**
161 * xmlErrInternal:
162 * @ctxt: an XML parser context
163 * @msg: the error message
164 * @str: error informations
165 *
166 * Handle an internal error
167 */
168 static void
xmlErrInternal(xmlParserCtxtPtr ctxt,const char * msg,const xmlChar * str)169 xmlErrInternal(xmlParserCtxtPtr ctxt, const char *msg, const xmlChar * str)
170 {
171 if ((ctxt != NULL) && (ctxt->disableSAX != 0) &&
172 (ctxt->instate == XML_PARSER_EOF))
173 return;
174 if (ctxt != NULL)
175 ctxt->errNo = XML_ERR_INTERNAL_ERROR;
176 __xmlRaiseError(NULL, NULL, NULL,
177 ctxt, NULL, XML_FROM_PARSER, XML_ERR_INTERNAL_ERROR,
178 XML_ERR_FATAL, NULL, 0, (const char *) str, NULL, NULL,
179 0, 0, msg, str);
180 if (ctxt != NULL) {
181 ctxt->wellFormed = 0;
182 if (ctxt->recovery == 0)
183 ctxt->disableSAX = 1;
184 }
185 }
186
187 /**
188 * xmlErrEncodingInt:
189 * @ctxt: an XML parser context
190 * @error: the error number
191 * @msg: the error message
192 * @val: an integer value
193 *
194 * n encoding error
195 */
196 static void
xmlErrEncodingInt(xmlParserCtxtPtr ctxt,xmlParserErrors error,const char * msg,int val)197 xmlErrEncodingInt(xmlParserCtxtPtr ctxt, xmlParserErrors error,
198 const char *msg, int val)
199 {
200 if ((ctxt != NULL) && (ctxt->disableSAX != 0) &&
201 (ctxt->instate == XML_PARSER_EOF))
202 return;
203 if (ctxt != NULL)
204 ctxt->errNo = error;
205 __xmlRaiseError(NULL, NULL, NULL,
206 ctxt, NULL, XML_FROM_PARSER, error, XML_ERR_FATAL,
207 NULL, 0, NULL, NULL, NULL, val, 0, msg, val);
208 if (ctxt != NULL) {
209 ctxt->wellFormed = 0;
210 if (ctxt->recovery == 0)
211 ctxt->disableSAX = 1;
212 }
213 }
214
215 /**
216 * xmlIsLetter:
217 * @c: an unicode character (int)
218 *
219 * Check whether the character is allowed by the production
220 * [84] Letter ::= BaseChar | Ideographic
221 *
222 * Returns 0 if not, non-zero otherwise
223 */
224 int
xmlIsLetter(int c)225 xmlIsLetter(int c) {
226 return(IS_BASECHAR(c) || IS_IDEOGRAPHIC(c));
227 }
228
229 /************************************************************************
230 * *
231 * Input handling functions for progressive parsing *
232 * *
233 ************************************************************************/
234
235 /* #define DEBUG_INPUT */
236 /* #define DEBUG_STACK */
237 /* #define DEBUG_PUSH */
238
239
240 /* we need to keep enough input to show errors in context */
241 #define LINE_LEN 80
242
243 #ifdef DEBUG_INPUT
244 #define CHECK_BUFFER(in) check_buffer(in)
245
246 static
check_buffer(xmlParserInputPtr in)247 void check_buffer(xmlParserInputPtr in) {
248 if (in->base != xmlBufContent(in->buf->buffer)) {
249 xmlGenericError(xmlGenericErrorContext,
250 "xmlParserInput: base mismatch problem\n");
251 }
252 if (in->cur < in->base) {
253 xmlGenericError(xmlGenericErrorContext,
254 "xmlParserInput: cur < base problem\n");
255 }
256 if (in->cur > in->base + xmlBufUse(in->buf->buffer)) {
257 xmlGenericError(xmlGenericErrorContext,
258 "xmlParserInput: cur > base + use problem\n");
259 }
260 xmlGenericError(xmlGenericErrorContext,"buffer %x : content %x, cur %d, use %d\n",
261 (int) in, (int) xmlBufContent(in->buf->buffer), in->cur - in->base,
262 xmlBufUse(in->buf->buffer));
263 }
264
265 #else
266 #define CHECK_BUFFER(in)
267 #endif
268
269
270 /**
271 * xmlParserInputRead:
272 * @in: an XML parser input
273 * @len: an indicative size for the lookahead
274 *
275 * This function was internal and is deprecated.
276 *
277 * Returns -1 as this is an error to use it.
278 */
279 int
xmlParserInputRead(xmlParserInputPtr in ATTRIBUTE_UNUSED,int len ATTRIBUTE_UNUSED)280 xmlParserInputRead(xmlParserInputPtr in ATTRIBUTE_UNUSED, int len ATTRIBUTE_UNUSED) {
281 return(-1);
282 }
283
284 /**
285 * xmlParserInputGrow:
286 * @in: an XML parser input
287 * @len: an indicative size for the lookahead
288 *
289 * This function increase the input for the parser. It tries to
290 * preserve pointers to the input buffer, and keep already read data
291 *
292 * Returns the amount of char read, or -1 in case of error, 0 indicate the
293 * end of this entity
294 */
295 int
xmlParserInputGrow(xmlParserInputPtr in,int len)296 xmlParserInputGrow(xmlParserInputPtr in, int len) {
297 size_t ret;
298 size_t indx;
299 const xmlChar *content;
300
301 if ((in == NULL) || (len < 0)) return(-1);
302 #ifdef DEBUG_INPUT
303 xmlGenericError(xmlGenericErrorContext, "Grow\n");
304 #endif
305 if (in->buf == NULL) return(-1);
306 if (in->base == NULL) return(-1);
307 if (in->cur == NULL) return(-1);
308 if (in->buf->buffer == NULL) return(-1);
309
310 CHECK_BUFFER(in);
311
312 indx = in->cur - in->base;
313 if (xmlBufUse(in->buf->buffer) > (unsigned int) indx + INPUT_CHUNK) {
314
315 CHECK_BUFFER(in);
316
317 return(0);
318 }
319 if (in->buf->readcallback != NULL) {
320 ret = xmlParserInputBufferGrow(in->buf, len);
321 } else
322 return(0);
323
324 /*
325 * NOTE : in->base may be a "dangling" i.e. freed pointer in this
326 * block, but we use it really as an integer to do some
327 * pointer arithmetic. Insure will raise it as a bug but in
328 * that specific case, that's not !
329 */
330
331 content = xmlBufContent(in->buf->buffer);
332 if (in->base != content) {
333 /*
334 * the buffer has been reallocated
335 */
336 indx = in->cur - in->base;
337 in->base = content;
338 in->cur = &content[indx];
339 }
340 in->end = xmlBufEnd(in->buf->buffer);
341
342 CHECK_BUFFER(in);
343
344 return(ret);
345 }
346
347 /**
348 * xmlParserInputShrink:
349 * @in: an XML parser input
350 *
351 * This function removes used input for the parser.
352 */
353 void
xmlParserInputShrink(xmlParserInputPtr in)354 xmlParserInputShrink(xmlParserInputPtr in) {
355 size_t used;
356 size_t ret;
357 size_t indx;
358 const xmlChar *content;
359
360 #ifdef DEBUG_INPUT
361 xmlGenericError(xmlGenericErrorContext, "Shrink\n");
362 #endif
363 if (in == NULL) return;
364 if (in->buf == NULL) return;
365 if (in->base == NULL) return;
366 if (in->cur == NULL) return;
367 if (in->buf->buffer == NULL) return;
368
369 CHECK_BUFFER(in);
370
371 used = in->cur - xmlBufContent(in->buf->buffer);
372 /*
373 * Do not shrink on large buffers whose only a tiny fraction
374 * was consumed
375 */
376 if (used > INPUT_CHUNK) {
377 ret = xmlBufShrink(in->buf->buffer, used - LINE_LEN);
378 if (ret > 0) {
379 in->cur -= ret;
380 in->consumed += ret;
381 }
382 in->end = xmlBufEnd(in->buf->buffer);
383 }
384
385 CHECK_BUFFER(in);
386
387 if (xmlBufUse(in->buf->buffer) > INPUT_CHUNK) {
388 return;
389 }
390 xmlParserInputBufferRead(in->buf, 2 * INPUT_CHUNK);
391 content = xmlBufContent(in->buf->buffer);
392 if (in->base != content) {
393 /*
394 * the buffer has been reallocated
395 */
396 indx = in->cur - in->base;
397 in->base = content;
398 in->cur = &content[indx];
399 }
400 in->end = xmlBufEnd(in->buf->buffer);
401
402 CHECK_BUFFER(in);
403 }
404
405 /************************************************************************
406 * *
407 * UTF8 character input and related functions *
408 * *
409 ************************************************************************/
410
411 /**
412 * xmlNextChar:
413 * @ctxt: the XML parser context
414 *
415 * Skip to the next char input char.
416 */
417
418 void
xmlNextChar(xmlParserCtxtPtr ctxt)419 xmlNextChar(xmlParserCtxtPtr ctxt)
420 {
421 if ((ctxt == NULL) || (ctxt->instate == XML_PARSER_EOF) ||
422 (ctxt->input == NULL))
423 return;
424
425 if (ctxt->charset == XML_CHAR_ENCODING_UTF8) {
426 if ((*ctxt->input->cur == 0) &&
427 (xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0) &&
428 (ctxt->instate != XML_PARSER_COMMENT)) {
429 /*
430 * If we are at the end of the current entity and
431 * the context allows it, we pop consumed entities
432 * automatically.
433 * the auto closing should be blocked in other cases
434 */
435 xmlPopInput(ctxt);
436 } else {
437 const unsigned char *cur;
438 unsigned char c;
439
440 /*
441 * 2.11 End-of-Line Handling
442 * the literal two-character sequence "#xD#xA" or a standalone
443 * literal #xD, an XML processor must pass to the application
444 * the single character #xA.
445 */
446 if (*(ctxt->input->cur) == '\n') {
447 ctxt->input->line++; ctxt->input->col = 1;
448 } else
449 ctxt->input->col++;
450
451 /*
452 * We are supposed to handle UTF8, check it's valid
453 * From rfc2044: encoding of the Unicode values on UTF-8:
454 *
455 * UCS-4 range (hex.) UTF-8 octet sequence (binary)
456 * 0000 0000-0000 007F 0xxxxxxx
457 * 0000 0080-0000 07FF 110xxxxx 10xxxxxx
458 * 0000 0800-0000 FFFF 1110xxxx 10xxxxxx 10xxxxxx
459 *
460 * Check for the 0x110000 limit too
461 */
462 cur = ctxt->input->cur;
463
464 c = *cur;
465 if (c & 0x80) {
466 if (c == 0xC0)
467 goto encoding_error;
468 if (cur[1] == 0) {
469 xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
470 cur = ctxt->input->cur;
471 }
472 if ((cur[1] & 0xc0) != 0x80)
473 goto encoding_error;
474 if ((c & 0xe0) == 0xe0) {
475 unsigned int val;
476
477 if (cur[2] == 0) {
478 xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
479 cur = ctxt->input->cur;
480 }
481 if ((cur[2] & 0xc0) != 0x80)
482 goto encoding_error;
483 if ((c & 0xf0) == 0xf0) {
484 if (cur[3] == 0) {
485 xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
486 cur = ctxt->input->cur;
487 }
488 if (((c & 0xf8) != 0xf0) ||
489 ((cur[3] & 0xc0) != 0x80))
490 goto encoding_error;
491 /* 4-byte code */
492 ctxt->input->cur += 4;
493 val = (cur[0] & 0x7) << 18;
494 val |= (cur[1] & 0x3f) << 12;
495 val |= (cur[2] & 0x3f) << 6;
496 val |= cur[3] & 0x3f;
497 } else {
498 /* 3-byte code */
499 ctxt->input->cur += 3;
500 val = (cur[0] & 0xf) << 12;
501 val |= (cur[1] & 0x3f) << 6;
502 val |= cur[2] & 0x3f;
503 }
504 if (((val > 0xd7ff) && (val < 0xe000)) ||
505 ((val > 0xfffd) && (val < 0x10000)) ||
506 (val >= 0x110000)) {
507 xmlErrEncodingInt(ctxt, XML_ERR_INVALID_CHAR,
508 "Char 0x%X out of allowed range\n",
509 val);
510 }
511 } else
512 /* 2-byte code */
513 ctxt->input->cur += 2;
514 } else
515 /* 1-byte code */
516 ctxt->input->cur++;
517
518 ctxt->nbChars++;
519 if (*ctxt->input->cur == 0)
520 xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
521 }
522 } else {
523 /*
524 * Assume it's a fixed length encoding (1) with
525 * a compatible encoding for the ASCII set, since
526 * XML constructs only use < 128 chars
527 */
528
529 if (*(ctxt->input->cur) == '\n') {
530 ctxt->input->line++; ctxt->input->col = 1;
531 } else
532 ctxt->input->col++;
533 ctxt->input->cur++;
534 ctxt->nbChars++;
535 if (*ctxt->input->cur == 0)
536 xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
537 }
538 if ((*ctxt->input->cur == '%') && (!ctxt->html))
539 xmlParserHandlePEReference(ctxt);
540 if ((*ctxt->input->cur == 0) &&
541 (xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0))
542 xmlPopInput(ctxt);
543 return;
544 encoding_error:
545 /*
546 * If we detect an UTF8 error that probably mean that the
547 * input encoding didn't get properly advertised in the
548 * declaration header. Report the error and switch the encoding
549 * to ISO-Latin-1 (if you don't like this policy, just declare the
550 * encoding !)
551 */
552 if ((ctxt == NULL) || (ctxt->input == NULL) ||
553 (ctxt->input->end - ctxt->input->cur < 4)) {
554 __xmlErrEncoding(ctxt, XML_ERR_INVALID_CHAR,
555 "Input is not proper UTF-8, indicate encoding !\n",
556 NULL, NULL);
557 } else {
558 char buffer[150];
559
560 snprintf(buffer, 149, "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n",
561 ctxt->input->cur[0], ctxt->input->cur[1],
562 ctxt->input->cur[2], ctxt->input->cur[3]);
563 __xmlErrEncoding(ctxt, XML_ERR_INVALID_CHAR,
564 "Input is not proper UTF-8, indicate encoding !\n%s",
565 BAD_CAST buffer, NULL);
566 }
567 ctxt->charset = XML_CHAR_ENCODING_8859_1;
568 ctxt->input->cur++;
569 return;
570 }
571
572 /**
573 * xmlCurrentChar:
574 * @ctxt: the XML parser context
575 * @len: pointer to the length of the char read
576 *
577 * The current char value, if using UTF-8 this may actually span multiple
578 * bytes in the input buffer. Implement the end of line normalization:
579 * 2.11 End-of-Line Handling
580 * Wherever an external parsed entity or the literal entity value
581 * of an internal parsed entity contains either the literal two-character
582 * sequence "#xD#xA" or a standalone literal #xD, an XML processor
583 * must pass to the application the single character #xA.
584 * This behavior can conveniently be produced by normalizing all
585 * line breaks to #xA on input, before parsing.)
586 *
587 * Returns the current char value and its length
588 */
589
590 int
xmlCurrentChar(xmlParserCtxtPtr ctxt,int * len)591 xmlCurrentChar(xmlParserCtxtPtr ctxt, int *len) {
592 if ((ctxt == NULL) || (len == NULL) || (ctxt->input == NULL)) return(0);
593 if (ctxt->instate == XML_PARSER_EOF)
594 return(0);
595
596 if ((*ctxt->input->cur >= 0x20) && (*ctxt->input->cur <= 0x7F)) {
597 *len = 1;
598 return((int) *ctxt->input->cur);
599 }
600 if (ctxt->charset == XML_CHAR_ENCODING_UTF8) {
601 /*
602 * We are supposed to handle UTF8, check it's valid
603 * From rfc2044: encoding of the Unicode values on UTF-8:
604 *
605 * UCS-4 range (hex.) UTF-8 octet sequence (binary)
606 * 0000 0000-0000 007F 0xxxxxxx
607 * 0000 0080-0000 07FF 110xxxxx 10xxxxxx
608 * 0000 0800-0000 FFFF 1110xxxx 10xxxxxx 10xxxxxx
609 *
610 * Check for the 0x110000 limit too
611 */
612 const unsigned char *cur = ctxt->input->cur;
613 unsigned char c;
614 unsigned int val;
615
616 c = *cur;
617 if (c & 0x80) {
618 if (((c & 0x40) == 0) || (c == 0xC0))
619 goto encoding_error;
620 if (cur[1] == 0) {
621 xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
622 cur = ctxt->input->cur;
623 }
624 if ((cur[1] & 0xc0) != 0x80)
625 goto encoding_error;
626 if ((c & 0xe0) == 0xe0) {
627 if (cur[2] == 0) {
628 xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
629 cur = ctxt->input->cur;
630 }
631 if ((cur[2] & 0xc0) != 0x80)
632 goto encoding_error;
633 if ((c & 0xf0) == 0xf0) {
634 if (cur[3] == 0) {
635 xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
636 cur = ctxt->input->cur;
637 }
638 if (((c & 0xf8) != 0xf0) ||
639 ((cur[3] & 0xc0) != 0x80))
640 goto encoding_error;
641 /* 4-byte code */
642 *len = 4;
643 val = (cur[0] & 0x7) << 18;
644 val |= (cur[1] & 0x3f) << 12;
645 val |= (cur[2] & 0x3f) << 6;
646 val |= cur[3] & 0x3f;
647 if (val < 0x10000)
648 goto encoding_error;
649 } else {
650 /* 3-byte code */
651 *len = 3;
652 val = (cur[0] & 0xf) << 12;
653 val |= (cur[1] & 0x3f) << 6;
654 val |= cur[2] & 0x3f;
655 if (val < 0x800)
656 goto encoding_error;
657 }
658 } else {
659 /* 2-byte code */
660 *len = 2;
661 val = (cur[0] & 0x1f) << 6;
662 val |= cur[1] & 0x3f;
663 if (val < 0x80)
664 goto encoding_error;
665 }
666 if (!IS_CHAR(val)) {
667 xmlErrEncodingInt(ctxt, XML_ERR_INVALID_CHAR,
668 "Char 0x%X out of allowed range\n", val);
669 }
670 return(val);
671 } else {
672 /* 1-byte code */
673 *len = 1;
674 if (*ctxt->input->cur == 0)
675 xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
676 if ((*ctxt->input->cur == 0) &&
677 (ctxt->input->end > ctxt->input->cur)) {
678 xmlErrEncodingInt(ctxt, XML_ERR_INVALID_CHAR,
679 "Char 0x0 out of allowed range\n", 0);
680 }
681 if (*ctxt->input->cur == 0xD) {
682 if (ctxt->input->cur[1] == 0xA) {
683 ctxt->nbChars++;
684 ctxt->input->cur++;
685 }
686 return(0xA);
687 }
688 return((int) *ctxt->input->cur);
689 }
690 }
691 /*
692 * Assume it's a fixed length encoding (1) with
693 * a compatible encoding for the ASCII set, since
694 * XML constructs only use < 128 chars
695 */
696 *len = 1;
697 if (*ctxt->input->cur == 0xD) {
698 if (ctxt->input->cur[1] == 0xA) {
699 ctxt->nbChars++;
700 ctxt->input->cur++;
701 }
702 return(0xA);
703 }
704 return((int) *ctxt->input->cur);
705 encoding_error:
706 /*
707 * An encoding problem may arise from a truncated input buffer
708 * splitting a character in the middle. In that case do not raise
709 * an error but return 0 to endicate an end of stream problem
710 */
711 if (ctxt->input->end - ctxt->input->cur < 4) {
712 *len = 0;
713 return(0);
714 }
715
716 /*
717 * If we detect an UTF8 error that probably mean that the
718 * input encoding didn't get properly advertised in the
719 * declaration header. Report the error and switch the encoding
720 * to ISO-Latin-1 (if you don't like this policy, just declare the
721 * encoding !)
722 */
723 {
724 char buffer[150];
725
726 snprintf(&buffer[0], 149, "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n",
727 ctxt->input->cur[0], ctxt->input->cur[1],
728 ctxt->input->cur[2], ctxt->input->cur[3]);
729 __xmlErrEncoding(ctxt, XML_ERR_INVALID_CHAR,
730 "Input is not proper UTF-8, indicate encoding !\n%s",
731 BAD_CAST buffer, NULL);
732 }
733 ctxt->charset = XML_CHAR_ENCODING_8859_1;
734 *len = 1;
735 return((int) *ctxt->input->cur);
736 }
737
738 /**
739 * xmlStringCurrentChar:
740 * @ctxt: the XML parser context
741 * @cur: pointer to the beginning of the char
742 * @len: pointer to the length of the char read
743 *
744 * The current char value, if using UTF-8 this may actually span multiple
745 * bytes in the input buffer.
746 *
747 * Returns the current char value and its length
748 */
749
750 int
xmlStringCurrentChar(xmlParserCtxtPtr ctxt,const xmlChar * cur,int * len)751 xmlStringCurrentChar(xmlParserCtxtPtr ctxt, const xmlChar * cur, int *len)
752 {
753 if ((len == NULL) || (cur == NULL)) return(0);
754 if ((ctxt == NULL) || (ctxt->charset == XML_CHAR_ENCODING_UTF8)) {
755 /*
756 * We are supposed to handle UTF8, check it's valid
757 * From rfc2044: encoding of the Unicode values on UTF-8:
758 *
759 * UCS-4 range (hex.) UTF-8 octet sequence (binary)
760 * 0000 0000-0000 007F 0xxxxxxx
761 * 0000 0080-0000 07FF 110xxxxx 10xxxxxx
762 * 0000 0800-0000 FFFF 1110xxxx 10xxxxxx 10xxxxxx
763 *
764 * Check for the 0x110000 limit too
765 */
766 unsigned char c;
767 unsigned int val;
768
769 c = *cur;
770 if (c & 0x80) {
771 if ((cur[1] & 0xc0) != 0x80)
772 goto encoding_error;
773 if ((c & 0xe0) == 0xe0) {
774
775 if ((cur[2] & 0xc0) != 0x80)
776 goto encoding_error;
777 if ((c & 0xf0) == 0xf0) {
778 if (((c & 0xf8) != 0xf0) || ((cur[3] & 0xc0) != 0x80))
779 goto encoding_error;
780 /* 4-byte code */
781 *len = 4;
782 val = (cur[0] & 0x7) << 18;
783 val |= (cur[1] & 0x3f) << 12;
784 val |= (cur[2] & 0x3f) << 6;
785 val |= cur[3] & 0x3f;
786 } else {
787 /* 3-byte code */
788 *len = 3;
789 val = (cur[0] & 0xf) << 12;
790 val |= (cur[1] & 0x3f) << 6;
791 val |= cur[2] & 0x3f;
792 }
793 } else {
794 /* 2-byte code */
795 *len = 2;
796 val = (cur[0] & 0x1f) << 6;
797 val |= cur[1] & 0x3f;
798 }
799 if (!IS_CHAR(val)) {
800 xmlErrEncodingInt(ctxt, XML_ERR_INVALID_CHAR,
801 "Char 0x%X out of allowed range\n", val);
802 }
803 return (val);
804 } else {
805 /* 1-byte code */
806 *len = 1;
807 return ((int) *cur);
808 }
809 }
810 /*
811 * Assume it's a fixed length encoding (1) with
812 * a compatible encoding for the ASCII set, since
813 * XML constructs only use < 128 chars
814 */
815 *len = 1;
816 return ((int) *cur);
817 encoding_error:
818
819 /*
820 * An encoding problem may arise from a truncated input buffer
821 * splitting a character in the middle. In that case do not raise
822 * an error but return 0 to endicate an end of stream problem
823 */
824 if ((ctxt == NULL) || (ctxt->input == NULL) ||
825 (ctxt->input->end - ctxt->input->cur < 4)) {
826 *len = 0;
827 return(0);
828 }
829 /*
830 * If we detect an UTF8 error that probably mean that the
831 * input encoding didn't get properly advertised in the
832 * declaration header. Report the error and switch the encoding
833 * to ISO-Latin-1 (if you don't like this policy, just declare the
834 * encoding !)
835 */
836 {
837 char buffer[150];
838
839 snprintf(buffer, 149, "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n",
840 ctxt->input->cur[0], ctxt->input->cur[1],
841 ctxt->input->cur[2], ctxt->input->cur[3]);
842 __xmlErrEncoding(ctxt, XML_ERR_INVALID_CHAR,
843 "Input is not proper UTF-8, indicate encoding !\n%s",
844 BAD_CAST buffer, NULL);
845 }
846 *len = 1;
847 return ((int) *cur);
848 }
849
850 /**
851 * xmlCopyCharMultiByte:
852 * @out: pointer to an array of xmlChar
853 * @val: the char value
854 *
855 * append the char value in the array
856 *
857 * Returns the number of xmlChar written
858 */
859 int
xmlCopyCharMultiByte(xmlChar * out,int val)860 xmlCopyCharMultiByte(xmlChar *out, int val) {
861 if (out == NULL) return(0);
862 /*
863 * We are supposed to handle UTF8, check it's valid
864 * From rfc2044: encoding of the Unicode values on UTF-8:
865 *
866 * UCS-4 range (hex.) UTF-8 octet sequence (binary)
867 * 0000 0000-0000 007F 0xxxxxxx
868 * 0000 0080-0000 07FF 110xxxxx 10xxxxxx
869 * 0000 0800-0000 FFFF 1110xxxx 10xxxxxx 10xxxxxx
870 */
871 if (val >= 0x80) {
872 xmlChar *savedout = out;
873 int bits;
874 if (val < 0x800) { *out++= (val >> 6) | 0xC0; bits= 0; }
875 else if (val < 0x10000) { *out++= (val >> 12) | 0xE0; bits= 6;}
876 else if (val < 0x110000) { *out++= (val >> 18) | 0xF0; bits= 12; }
877 else {
878 xmlErrEncodingInt(NULL, XML_ERR_INVALID_CHAR,
879 "Internal error, xmlCopyCharMultiByte 0x%X out of bound\n",
880 val);
881 return(0);
882 }
883 for ( ; bits >= 0; bits-= 6)
884 *out++= ((val >> bits) & 0x3F) | 0x80 ;
885 return (out - savedout);
886 }
887 *out = (xmlChar) val;
888 return 1;
889 }
890
891 /**
892 * xmlCopyChar:
893 * @len: Ignored, compatibility
894 * @out: pointer to an array of xmlChar
895 * @val: the char value
896 *
897 * append the char value in the array
898 *
899 * Returns the number of xmlChar written
900 */
901
902 int
xmlCopyChar(int len ATTRIBUTE_UNUSED,xmlChar * out,int val)903 xmlCopyChar(int len ATTRIBUTE_UNUSED, xmlChar *out, int val) {
904 if (out == NULL) return(0);
905 /* the len parameter is ignored */
906 if (val >= 0x80) {
907 return(xmlCopyCharMultiByte (out, val));
908 }
909 *out = (xmlChar) val;
910 return 1;
911 }
912
913 /************************************************************************
914 * *
915 * Commodity functions to switch encodings *
916 * *
917 ************************************************************************/
918
919 static int
920 xmlSwitchToEncodingInt(xmlParserCtxtPtr ctxt,
921 xmlCharEncodingHandlerPtr handler, int len);
922 static int
923 xmlSwitchInputEncodingInt(xmlParserCtxtPtr ctxt, xmlParserInputPtr input,
924 xmlCharEncodingHandlerPtr handler, int len);
925 /**
926 * xmlSwitchEncoding:
927 * @ctxt: the parser context
928 * @enc: the encoding value (number)
929 *
930 * change the input functions when discovering the character encoding
931 * of a given entity.
932 *
933 * Returns 0 in case of success, -1 otherwise
934 */
935 int
xmlSwitchEncoding(xmlParserCtxtPtr ctxt,xmlCharEncoding enc)936 xmlSwitchEncoding(xmlParserCtxtPtr ctxt, xmlCharEncoding enc)
937 {
938 xmlCharEncodingHandlerPtr handler;
939 int len = -1;
940
941 if (ctxt == NULL) return(-1);
942 switch (enc) {
943 case XML_CHAR_ENCODING_ERROR:
944 __xmlErrEncoding(ctxt, XML_ERR_UNKNOWN_ENCODING,
945 "encoding unknown\n", NULL, NULL);
946 return(-1);
947 case XML_CHAR_ENCODING_NONE:
948 /* let's assume it's UTF-8 without the XML decl */
949 ctxt->charset = XML_CHAR_ENCODING_UTF8;
950 return(0);
951 case XML_CHAR_ENCODING_UTF8:
952 /* default encoding, no conversion should be needed */
953 ctxt->charset = XML_CHAR_ENCODING_UTF8;
954
955 /*
956 * Errata on XML-1.0 June 20 2001
957 * Specific handling of the Byte Order Mark for
958 * UTF-8
959 */
960 if ((ctxt->input != NULL) &&
961 (ctxt->input->cur[0] == 0xEF) &&
962 (ctxt->input->cur[1] == 0xBB) &&
963 (ctxt->input->cur[2] == 0xBF)) {
964 ctxt->input->cur += 3;
965 }
966 return(0);
967 case XML_CHAR_ENCODING_UTF16LE:
968 case XML_CHAR_ENCODING_UTF16BE:
969 /*The raw input characters are encoded
970 *in UTF-16. As we expect this function
971 *to be called after xmlCharEncInFunc, we expect
972 *ctxt->input->cur to contain UTF-8 encoded characters.
973 *So the raw UTF16 Byte Order Mark
974 *has also been converted into
975 *an UTF-8 BOM. Let's skip that BOM.
976 */
977 if ((ctxt->input != NULL) && (ctxt->input->cur != NULL) &&
978 (ctxt->input->cur[0] == 0xEF) &&
979 (ctxt->input->cur[1] == 0xBB) &&
980 (ctxt->input->cur[2] == 0xBF)) {
981 ctxt->input->cur += 3;
982 }
983 len = 90;
984 break;
985 case XML_CHAR_ENCODING_UCS2:
986 len = 90;
987 break;
988 case XML_CHAR_ENCODING_UCS4BE:
989 case XML_CHAR_ENCODING_UCS4LE:
990 case XML_CHAR_ENCODING_UCS4_2143:
991 case XML_CHAR_ENCODING_UCS4_3412:
992 len = 180;
993 break;
994 case XML_CHAR_ENCODING_EBCDIC:
995 case XML_CHAR_ENCODING_8859_1:
996 case XML_CHAR_ENCODING_8859_2:
997 case XML_CHAR_ENCODING_8859_3:
998 case XML_CHAR_ENCODING_8859_4:
999 case XML_CHAR_ENCODING_8859_5:
1000 case XML_CHAR_ENCODING_8859_6:
1001 case XML_CHAR_ENCODING_8859_7:
1002 case XML_CHAR_ENCODING_8859_8:
1003 case XML_CHAR_ENCODING_8859_9:
1004 case XML_CHAR_ENCODING_ASCII:
1005 case XML_CHAR_ENCODING_2022_JP:
1006 case XML_CHAR_ENCODING_SHIFT_JIS:
1007 case XML_CHAR_ENCODING_EUC_JP:
1008 len = 45;
1009 break;
1010 }
1011 handler = xmlGetCharEncodingHandler(enc);
1012 if (handler == NULL) {
1013 /*
1014 * Default handlers.
1015 */
1016 switch (enc) {
1017 case XML_CHAR_ENCODING_ASCII:
1018 /* default encoding, no conversion should be needed */
1019 ctxt->charset = XML_CHAR_ENCODING_UTF8;
1020 return(0);
1021 case XML_CHAR_ENCODING_UTF16LE:
1022 break;
1023 case XML_CHAR_ENCODING_UTF16BE:
1024 break;
1025 case XML_CHAR_ENCODING_UCS4LE:
1026 __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
1027 "encoding not supported %s\n",
1028 BAD_CAST "USC4 little endian", NULL);
1029 break;
1030 case XML_CHAR_ENCODING_UCS4BE:
1031 __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
1032 "encoding not supported %s\n",
1033 BAD_CAST "USC4 big endian", NULL);
1034 break;
1035 case XML_CHAR_ENCODING_EBCDIC:
1036 __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
1037 "encoding not supported %s\n",
1038 BAD_CAST "EBCDIC", NULL);
1039 break;
1040 case XML_CHAR_ENCODING_UCS4_2143:
1041 __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
1042 "encoding not supported %s\n",
1043 BAD_CAST "UCS4 2143", NULL);
1044 break;
1045 case XML_CHAR_ENCODING_UCS4_3412:
1046 __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
1047 "encoding not supported %s\n",
1048 BAD_CAST "UCS4 3412", NULL);
1049 break;
1050 case XML_CHAR_ENCODING_UCS2:
1051 __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
1052 "encoding not supported %s\n",
1053 BAD_CAST "UCS2", NULL);
1054 break;
1055 case XML_CHAR_ENCODING_8859_1:
1056 case XML_CHAR_ENCODING_8859_2:
1057 case XML_CHAR_ENCODING_8859_3:
1058 case XML_CHAR_ENCODING_8859_4:
1059 case XML_CHAR_ENCODING_8859_5:
1060 case XML_CHAR_ENCODING_8859_6:
1061 case XML_CHAR_ENCODING_8859_7:
1062 case XML_CHAR_ENCODING_8859_8:
1063 case XML_CHAR_ENCODING_8859_9:
1064 /*
1065 * We used to keep the internal content in the
1066 * document encoding however this turns being unmaintainable
1067 * So xmlGetCharEncodingHandler() will return non-null
1068 * values for this now.
1069 */
1070 if ((ctxt->inputNr == 1) &&
1071 (ctxt->encoding == NULL) &&
1072 (ctxt->input != NULL) &&
1073 (ctxt->input->encoding != NULL)) {
1074 ctxt->encoding = xmlStrdup(ctxt->input->encoding);
1075 }
1076 ctxt->charset = enc;
1077 return(0);
1078 case XML_CHAR_ENCODING_2022_JP:
1079 __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
1080 "encoding not supported %s\n",
1081 BAD_CAST "ISO-2022-JP", NULL);
1082 break;
1083 case XML_CHAR_ENCODING_SHIFT_JIS:
1084 __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
1085 "encoding not supported %s\n",
1086 BAD_CAST "Shift_JIS", NULL);
1087 break;
1088 case XML_CHAR_ENCODING_EUC_JP:
1089 __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
1090 "encoding not supported %s\n",
1091 BAD_CAST "EUC-JP", NULL);
1092 break;
1093 default:
1094 break;
1095 }
1096 }
1097 if (handler == NULL)
1098 return(-1);
1099 ctxt->charset = XML_CHAR_ENCODING_UTF8;
1100 return(xmlSwitchToEncodingInt(ctxt, handler, len));
1101 }
1102
1103 /**
1104 * xmlSwitchInputEncoding:
1105 * @ctxt: the parser context
1106 * @input: the input stream
1107 * @handler: the encoding handler
1108 * @len: the number of bytes to convert for the first line or -1
1109 *
1110 * change the input functions when discovering the character encoding
1111 * of a given entity.
1112 *
1113 * Returns 0 in case of success, -1 otherwise
1114 */
1115 static int
xmlSwitchInputEncodingInt(xmlParserCtxtPtr ctxt,xmlParserInputPtr input,xmlCharEncodingHandlerPtr handler,int len)1116 xmlSwitchInputEncodingInt(xmlParserCtxtPtr ctxt, xmlParserInputPtr input,
1117 xmlCharEncodingHandlerPtr handler, int len)
1118 {
1119 int nbchars;
1120
1121 if (handler == NULL)
1122 return (-1);
1123 if (input == NULL)
1124 return (-1);
1125 if (input->buf != NULL) {
1126 if (input->buf->encoder != NULL) {
1127 /*
1128 * Check in case the auto encoding detetection triggered
1129 * in already.
1130 */
1131 if (input->buf->encoder == handler)
1132 return (0);
1133
1134 /*
1135 * "UTF-16" can be used for both LE and BE
1136 if ((!xmlStrncmp(BAD_CAST input->buf->encoder->name,
1137 BAD_CAST "UTF-16", 6)) &&
1138 (!xmlStrncmp(BAD_CAST handler->name,
1139 BAD_CAST "UTF-16", 6))) {
1140 return(0);
1141 }
1142 */
1143
1144 /*
1145 * Note: this is a bit dangerous, but that's what it
1146 * takes to use nearly compatible signature for different
1147 * encodings.
1148 */
1149 xmlCharEncCloseFunc(input->buf->encoder);
1150 input->buf->encoder = handler;
1151 return (0);
1152 }
1153 input->buf->encoder = handler;
1154
1155 /*
1156 * Is there already some content down the pipe to convert ?
1157 */
1158 if (xmlBufIsEmpty(input->buf->buffer) == 0) {
1159 int processed;
1160 unsigned int use;
1161
1162 /*
1163 * Specific handling of the Byte Order Mark for
1164 * UTF-16
1165 */
1166 if ((handler->name != NULL) &&
1167 (!strcmp(handler->name, "UTF-16LE") ||
1168 !strcmp(handler->name, "UTF-16")) &&
1169 (input->cur[0] == 0xFF) && (input->cur[1] == 0xFE)) {
1170 input->cur += 2;
1171 }
1172 if ((handler->name != NULL) &&
1173 (!strcmp(handler->name, "UTF-16BE")) &&
1174 (input->cur[0] == 0xFE) && (input->cur[1] == 0xFF)) {
1175 input->cur += 2;
1176 }
1177 /*
1178 * Errata on XML-1.0 June 20 2001
1179 * Specific handling of the Byte Order Mark for
1180 * UTF-8
1181 */
1182 if ((handler->name != NULL) &&
1183 (!strcmp(handler->name, "UTF-8")) &&
1184 (input->cur[0] == 0xEF) &&
1185 (input->cur[1] == 0xBB) && (input->cur[2] == 0xBF)) {
1186 input->cur += 3;
1187 }
1188
1189 /*
1190 * Shrink the current input buffer.
1191 * Move it as the raw buffer and create a new input buffer
1192 */
1193 processed = input->cur - input->base;
1194 xmlBufShrink(input->buf->buffer, processed);
1195 input->buf->raw = input->buf->buffer;
1196 input->buf->buffer = xmlBufCreate();
1197 input->buf->rawconsumed = processed;
1198 use = xmlBufUse(input->buf->raw);
1199
1200 if (ctxt->html) {
1201 /*
1202 * convert as much as possible of the buffer
1203 */
1204 nbchars = xmlCharEncInput(input->buf, 1);
1205 } else {
1206 /*
1207 * convert just enough to get
1208 * '<?xml version="1.0" encoding="xxx"?>'
1209 * parsed with the autodetected encoding
1210 * into the parser reading buffer.
1211 */
1212 nbchars = xmlCharEncFirstLineInput(input->buf, len);
1213 }
1214 if (nbchars < 0) {
1215 xmlErrInternal(ctxt,
1216 "switching encoding: encoder error\n",
1217 NULL);
1218 return (-1);
1219 }
1220 input->buf->rawconsumed += use - xmlBufUse(input->buf->raw);
1221 xmlBufResetInput(input->buf->buffer, input);
1222 }
1223 return (0);
1224 } else if (input->length == 0) {
1225 /*
1226 * When parsing a static memory array one must know the
1227 * size to be able to convert the buffer.
1228 */
1229 xmlErrInternal(ctxt, "switching encoding : no input\n", NULL);
1230 return (-1);
1231 }
1232 return (0);
1233 }
1234
1235 /**
1236 * xmlSwitchInputEncoding:
1237 * @ctxt: the parser context
1238 * @input: the input stream
1239 * @handler: the encoding handler
1240 *
1241 * change the input functions when discovering the character encoding
1242 * of a given entity.
1243 *
1244 * Returns 0 in case of success, -1 otherwise
1245 */
1246 int
xmlSwitchInputEncoding(xmlParserCtxtPtr ctxt,xmlParserInputPtr input,xmlCharEncodingHandlerPtr handler)1247 xmlSwitchInputEncoding(xmlParserCtxtPtr ctxt, xmlParserInputPtr input,
1248 xmlCharEncodingHandlerPtr handler) {
1249 return(xmlSwitchInputEncodingInt(ctxt, input, handler, -1));
1250 }
1251
1252 /**
1253 * xmlSwitchToEncodingInt:
1254 * @ctxt: the parser context
1255 * @handler: the encoding handler
1256 * @len: the length to convert or -1
1257 *
1258 * change the input functions when discovering the character encoding
1259 * of a given entity, and convert only @len bytes of the output, this
1260 * is needed on auto detect to allows any declared encoding later to
1261 * convert the actual content after the xmlDecl
1262 *
1263 * Returns 0 in case of success, -1 otherwise
1264 */
1265 static int
xmlSwitchToEncodingInt(xmlParserCtxtPtr ctxt,xmlCharEncodingHandlerPtr handler,int len)1266 xmlSwitchToEncodingInt(xmlParserCtxtPtr ctxt,
1267 xmlCharEncodingHandlerPtr handler, int len) {
1268 int ret = 0;
1269
1270 if (handler != NULL) {
1271 if (ctxt->input != NULL) {
1272 ret = xmlSwitchInputEncodingInt(ctxt, ctxt->input, handler, len);
1273 } else {
1274 xmlErrInternal(ctxt, "xmlSwitchToEncoding : no input\n",
1275 NULL);
1276 return(-1);
1277 }
1278 /*
1279 * The parsing is now done in UTF8 natively
1280 */
1281 ctxt->charset = XML_CHAR_ENCODING_UTF8;
1282 } else
1283 return(-1);
1284 return(ret);
1285 }
1286
1287 /**
1288 * xmlSwitchToEncoding:
1289 * @ctxt: the parser context
1290 * @handler: the encoding handler
1291 *
1292 * change the input functions when discovering the character encoding
1293 * of a given entity.
1294 *
1295 * Returns 0 in case of success, -1 otherwise
1296 */
1297 int
xmlSwitchToEncoding(xmlParserCtxtPtr ctxt,xmlCharEncodingHandlerPtr handler)1298 xmlSwitchToEncoding(xmlParserCtxtPtr ctxt, xmlCharEncodingHandlerPtr handler)
1299 {
1300 return (xmlSwitchToEncodingInt(ctxt, handler, -1));
1301 }
1302
1303 /************************************************************************
1304 * *
1305 * Commodity functions to handle entities processing *
1306 * *
1307 ************************************************************************/
1308
1309 /**
1310 * xmlFreeInputStream:
1311 * @input: an xmlParserInputPtr
1312 *
1313 * Free up an input stream.
1314 */
1315 void
xmlFreeInputStream(xmlParserInputPtr input)1316 xmlFreeInputStream(xmlParserInputPtr input) {
1317 if (input == NULL) return;
1318
1319 if (input->filename != NULL) xmlFree((char *) input->filename);
1320 if (input->directory != NULL) xmlFree((char *) input->directory);
1321 if (input->encoding != NULL) xmlFree((char *) input->encoding);
1322 if (input->version != NULL) xmlFree((char *) input->version);
1323 if ((input->free != NULL) && (input->base != NULL))
1324 input->free((xmlChar *) input->base);
1325 if (input->buf != NULL)
1326 xmlFreeParserInputBuffer(input->buf);
1327 xmlFree(input);
1328 }
1329
1330 /**
1331 * xmlNewInputStream:
1332 * @ctxt: an XML parser context
1333 *
1334 * Create a new input stream structure.
1335 *
1336 * Returns the new input stream or NULL
1337 */
1338 xmlParserInputPtr
xmlNewInputStream(xmlParserCtxtPtr ctxt)1339 xmlNewInputStream(xmlParserCtxtPtr ctxt) {
1340 xmlParserInputPtr input;
1341
1342 input = (xmlParserInputPtr) xmlMalloc(sizeof(xmlParserInput));
1343 if (input == NULL) {
1344 xmlErrMemory(ctxt, "couldn't allocate a new input stream\n");
1345 return(NULL);
1346 }
1347 memset(input, 0, sizeof(xmlParserInput));
1348 input->line = 1;
1349 input->col = 1;
1350 input->standalone = -1;
1351
1352 /*
1353 * If the context is NULL the id cannot be initialized, but that
1354 * should not happen while parsing which is the situation where
1355 * the id is actually needed.
1356 */
1357 if (ctxt != NULL)
1358 input->id = ctxt->input_id++;
1359
1360 return(input);
1361 }
1362
1363 /**
1364 * xmlNewIOInputStream:
1365 * @ctxt: an XML parser context
1366 * @input: an I/O Input
1367 * @enc: the charset encoding if known
1368 *
1369 * Create a new input stream structure encapsulating the @input into
1370 * a stream suitable for the parser.
1371 *
1372 * Returns the new input stream or NULL
1373 */
1374 xmlParserInputPtr
xmlNewIOInputStream(xmlParserCtxtPtr ctxt,xmlParserInputBufferPtr input,xmlCharEncoding enc)1375 xmlNewIOInputStream(xmlParserCtxtPtr ctxt, xmlParserInputBufferPtr input,
1376 xmlCharEncoding enc) {
1377 xmlParserInputPtr inputStream;
1378
1379 if (input == NULL) return(NULL);
1380 if (xmlParserDebugEntities)
1381 xmlGenericError(xmlGenericErrorContext, "new input from I/O\n");
1382 inputStream = xmlNewInputStream(ctxt);
1383 if (inputStream == NULL) {
1384 return(NULL);
1385 }
1386 inputStream->filename = NULL;
1387 inputStream->buf = input;
1388 xmlBufResetInput(inputStream->buf->buffer, inputStream);
1389
1390 if (enc != XML_CHAR_ENCODING_NONE) {
1391 xmlSwitchEncoding(ctxt, enc);
1392 }
1393
1394 return(inputStream);
1395 }
1396
1397 /**
1398 * xmlNewEntityInputStream:
1399 * @ctxt: an XML parser context
1400 * @entity: an Entity pointer
1401 *
1402 * Create a new input stream based on an xmlEntityPtr
1403 *
1404 * Returns the new input stream or NULL
1405 */
1406 xmlParserInputPtr
xmlNewEntityInputStream(xmlParserCtxtPtr ctxt,xmlEntityPtr entity)1407 xmlNewEntityInputStream(xmlParserCtxtPtr ctxt, xmlEntityPtr entity) {
1408 xmlParserInputPtr input;
1409
1410 if (entity == NULL) {
1411 xmlErrInternal(ctxt, "xmlNewEntityInputStream entity = NULL\n",
1412 NULL);
1413 return(NULL);
1414 }
1415 if (xmlParserDebugEntities)
1416 xmlGenericError(xmlGenericErrorContext,
1417 "new input from entity: %s\n", entity->name);
1418 if (entity->content == NULL) {
1419 switch (entity->etype) {
1420 case XML_EXTERNAL_GENERAL_UNPARSED_ENTITY:
1421 xmlErrInternal(ctxt, "Cannot parse entity %s\n",
1422 entity->name);
1423 break;
1424 case XML_EXTERNAL_GENERAL_PARSED_ENTITY:
1425 case XML_EXTERNAL_PARAMETER_ENTITY:
1426 return(xmlLoadExternalEntity((char *) entity->URI,
1427 (char *) entity->ExternalID, ctxt));
1428 case XML_INTERNAL_GENERAL_ENTITY:
1429 xmlErrInternal(ctxt,
1430 "Internal entity %s without content !\n",
1431 entity->name);
1432 break;
1433 case XML_INTERNAL_PARAMETER_ENTITY:
1434 xmlErrInternal(ctxt,
1435 "Internal parameter entity %s without content !\n",
1436 entity->name);
1437 break;
1438 case XML_INTERNAL_PREDEFINED_ENTITY:
1439 xmlErrInternal(ctxt,
1440 "Predefined entity %s without content !\n",
1441 entity->name);
1442 break;
1443 }
1444 return(NULL);
1445 }
1446 input = xmlNewInputStream(ctxt);
1447 if (input == NULL) {
1448 return(NULL);
1449 }
1450 if (entity->URI != NULL)
1451 input->filename = (char *) xmlStrdup((xmlChar *) entity->URI);
1452 input->base = entity->content;
1453 input->cur = entity->content;
1454 input->length = entity->length;
1455 input->end = &entity->content[input->length];
1456 return(input);
1457 }
1458
1459 /**
1460 * xmlNewStringInputStream:
1461 * @ctxt: an XML parser context
1462 * @buffer: an memory buffer
1463 *
1464 * Create a new input stream based on a memory buffer.
1465 * Returns the new input stream
1466 */
1467 xmlParserInputPtr
xmlNewStringInputStream(xmlParserCtxtPtr ctxt,const xmlChar * buffer)1468 xmlNewStringInputStream(xmlParserCtxtPtr ctxt, const xmlChar *buffer) {
1469 xmlParserInputPtr input;
1470
1471 if (buffer == NULL) {
1472 xmlErrInternal(ctxt, "xmlNewStringInputStream string = NULL\n",
1473 NULL);
1474 return(NULL);
1475 }
1476 if (xmlParserDebugEntities)
1477 xmlGenericError(xmlGenericErrorContext,
1478 "new fixed input: %.30s\n", buffer);
1479 input = xmlNewInputStream(ctxt);
1480 if (input == NULL) {
1481 xmlErrMemory(ctxt, "couldn't allocate a new input stream\n");
1482 return(NULL);
1483 }
1484 input->base = buffer;
1485 input->cur = buffer;
1486 input->length = xmlStrlen(buffer);
1487 input->end = &buffer[input->length];
1488 return(input);
1489 }
1490
1491 /**
1492 * xmlNewInputFromFile:
1493 * @ctxt: an XML parser context
1494 * @filename: the filename to use as entity
1495 *
1496 * Create a new input stream based on a file or an URL.
1497 *
1498 * Returns the new input stream or NULL in case of error
1499 */
1500 xmlParserInputPtr
xmlNewInputFromFile(xmlParserCtxtPtr ctxt,const char * filename)1501 xmlNewInputFromFile(xmlParserCtxtPtr ctxt, const char *filename) {
1502 xmlParserInputBufferPtr buf;
1503 xmlParserInputPtr inputStream;
1504 char *directory = NULL;
1505 xmlChar *URI = NULL;
1506
1507 if (xmlParserDebugEntities)
1508 xmlGenericError(xmlGenericErrorContext,
1509 "new input from file: %s\n", filename);
1510 if (ctxt == NULL) return(NULL);
1511 buf = xmlParserInputBufferCreateFilename(filename, XML_CHAR_ENCODING_NONE);
1512 if (buf == NULL) {
1513 if (filename == NULL)
1514 __xmlLoaderErr(ctxt,
1515 "failed to load external entity: NULL filename \n",
1516 NULL);
1517 else
1518 __xmlLoaderErr(ctxt, "failed to load external entity \"%s\"\n",
1519 (const char *) filename);
1520 return(NULL);
1521 }
1522
1523 inputStream = xmlNewInputStream(ctxt);
1524 if (inputStream == NULL)
1525 return(NULL);
1526
1527 inputStream->buf = buf;
1528 inputStream = xmlCheckHTTPInput(ctxt, inputStream);
1529 if (inputStream == NULL)
1530 return(NULL);
1531
1532 if (inputStream->filename == NULL)
1533 URI = xmlStrdup((xmlChar *) filename);
1534 else
1535 URI = xmlStrdup((xmlChar *) inputStream->filename);
1536 directory = xmlParserGetDirectory((const char *) URI);
1537 if (inputStream->filename != NULL) xmlFree((char *)inputStream->filename);
1538 inputStream->filename = (char *) xmlCanonicPath((const xmlChar *) URI);
1539 if (URI != NULL) xmlFree((char *) URI);
1540 inputStream->directory = directory;
1541
1542 xmlBufResetInput(inputStream->buf->buffer, inputStream);
1543 if ((ctxt->directory == NULL) && (directory != NULL))
1544 ctxt->directory = (char *) xmlStrdup((const xmlChar *) directory);
1545 return(inputStream);
1546 }
1547
1548 /************************************************************************
1549 * *
1550 * Commodity functions to handle parser contexts *
1551 * *
1552 ************************************************************************/
1553
1554 /**
1555 * xmlInitParserCtxt:
1556 * @ctxt: an XML parser context
1557 *
1558 * Initialize a parser context
1559 *
1560 * Returns 0 in case of success and -1 in case of error
1561 */
1562
1563 int
xmlInitParserCtxt(xmlParserCtxtPtr ctxt)1564 xmlInitParserCtxt(xmlParserCtxtPtr ctxt)
1565 {
1566 xmlParserInputPtr input;
1567
1568 if(ctxt==NULL) {
1569 xmlErrInternal(NULL, "Got NULL parser context\n", NULL);
1570 return(-1);
1571 }
1572
1573 xmlDefaultSAXHandlerInit();
1574
1575 if (ctxt->dict == NULL)
1576 ctxt->dict = xmlDictCreate();
1577 if (ctxt->dict == NULL) {
1578 xmlErrMemory(NULL, "cannot initialize parser context\n");
1579 return(-1);
1580 }
1581 xmlDictSetLimit(ctxt->dict, XML_MAX_DICTIONARY_LIMIT);
1582
1583 if (ctxt->sax == NULL)
1584 ctxt->sax = (xmlSAXHandler *) xmlMalloc(sizeof(xmlSAXHandler));
1585 if (ctxt->sax == NULL) {
1586 xmlErrMemory(NULL, "cannot initialize parser context\n");
1587 return(-1);
1588 }
1589 else
1590 xmlSAXVersion(ctxt->sax, 2);
1591
1592 ctxt->maxatts = 0;
1593 ctxt->atts = NULL;
1594 /* Allocate the Input stack */
1595 if (ctxt->inputTab == NULL) {
1596 ctxt->inputTab = (xmlParserInputPtr *)
1597 xmlMalloc(5 * sizeof(xmlParserInputPtr));
1598 ctxt->inputMax = 5;
1599 }
1600 if (ctxt->inputTab == NULL) {
1601 xmlErrMemory(NULL, "cannot initialize parser context\n");
1602 ctxt->inputNr = 0;
1603 ctxt->inputMax = 0;
1604 ctxt->input = NULL;
1605 return(-1);
1606 }
1607 while ((input = inputPop(ctxt)) != NULL) { /* Non consuming */
1608 xmlFreeInputStream(input);
1609 }
1610 ctxt->inputNr = 0;
1611 ctxt->input = NULL;
1612
1613 ctxt->version = NULL;
1614 ctxt->encoding = NULL;
1615 ctxt->standalone = -1;
1616 ctxt->hasExternalSubset = 0;
1617 ctxt->hasPErefs = 0;
1618 ctxt->html = 0;
1619 ctxt->external = 0;
1620 ctxt->instate = XML_PARSER_START;
1621 ctxt->token = 0;
1622 ctxt->directory = NULL;
1623
1624 /* Allocate the Node stack */
1625 if (ctxt->nodeTab == NULL) {
1626 ctxt->nodeTab = (xmlNodePtr *) xmlMalloc(10 * sizeof(xmlNodePtr));
1627 ctxt->nodeMax = 10;
1628 }
1629 if (ctxt->nodeTab == NULL) {
1630 xmlErrMemory(NULL, "cannot initialize parser context\n");
1631 ctxt->nodeNr = 0;
1632 ctxt->nodeMax = 0;
1633 ctxt->node = NULL;
1634 ctxt->inputNr = 0;
1635 ctxt->inputMax = 0;
1636 ctxt->input = NULL;
1637 return(-1);
1638 }
1639 ctxt->nodeNr = 0;
1640 ctxt->node = NULL;
1641
1642 /* Allocate the Name stack */
1643 if (ctxt->nameTab == NULL) {
1644 ctxt->nameTab = (const xmlChar **) xmlMalloc(10 * sizeof(xmlChar *));
1645 ctxt->nameMax = 10;
1646 }
1647 if (ctxt->nameTab == NULL) {
1648 xmlErrMemory(NULL, "cannot initialize parser context\n");
1649 ctxt->nodeNr = 0;
1650 ctxt->nodeMax = 0;
1651 ctxt->node = NULL;
1652 ctxt->inputNr = 0;
1653 ctxt->inputMax = 0;
1654 ctxt->input = NULL;
1655 ctxt->nameNr = 0;
1656 ctxt->nameMax = 0;
1657 ctxt->name = NULL;
1658 return(-1);
1659 }
1660 ctxt->nameNr = 0;
1661 ctxt->name = NULL;
1662
1663 /* Allocate the space stack */
1664 if (ctxt->spaceTab == NULL) {
1665 ctxt->spaceTab = (int *) xmlMalloc(10 * sizeof(int));
1666 ctxt->spaceMax = 10;
1667 }
1668 if (ctxt->spaceTab == NULL) {
1669 xmlErrMemory(NULL, "cannot initialize parser context\n");
1670 ctxt->nodeNr = 0;
1671 ctxt->nodeMax = 0;
1672 ctxt->node = NULL;
1673 ctxt->inputNr = 0;
1674 ctxt->inputMax = 0;
1675 ctxt->input = NULL;
1676 ctxt->nameNr = 0;
1677 ctxt->nameMax = 0;
1678 ctxt->name = NULL;
1679 ctxt->spaceNr = 0;
1680 ctxt->spaceMax = 0;
1681 ctxt->space = NULL;
1682 return(-1);
1683 }
1684 ctxt->spaceNr = 1;
1685 ctxt->spaceMax = 10;
1686 ctxt->spaceTab[0] = -1;
1687 ctxt->space = &ctxt->spaceTab[0];
1688 ctxt->userData = ctxt;
1689 ctxt->myDoc = NULL;
1690 ctxt->wellFormed = 1;
1691 ctxt->nsWellFormed = 1;
1692 ctxt->valid = 1;
1693 ctxt->loadsubset = xmlLoadExtDtdDefaultValue;
1694 if (ctxt->loadsubset) {
1695 ctxt->options |= XML_PARSE_DTDLOAD;
1696 }
1697 ctxt->validate = xmlDoValidityCheckingDefaultValue;
1698 ctxt->pedantic = xmlPedanticParserDefaultValue;
1699 if (ctxt->pedantic) {
1700 ctxt->options |= XML_PARSE_PEDANTIC;
1701 }
1702 ctxt->linenumbers = xmlLineNumbersDefaultValue;
1703 ctxt->keepBlanks = xmlKeepBlanksDefaultValue;
1704 if (ctxt->keepBlanks == 0) {
1705 ctxt->sax->ignorableWhitespace = xmlSAX2IgnorableWhitespace;
1706 ctxt->options |= XML_PARSE_NOBLANKS;
1707 }
1708
1709 ctxt->vctxt.finishDtd = XML_CTXT_FINISH_DTD_0;
1710 ctxt->vctxt.userData = ctxt;
1711 ctxt->vctxt.error = xmlParserValidityError;
1712 ctxt->vctxt.warning = xmlParserValidityWarning;
1713 if (ctxt->validate) {
1714 if (xmlGetWarningsDefaultValue == 0)
1715 ctxt->vctxt.warning = NULL;
1716 else
1717 ctxt->vctxt.warning = xmlParserValidityWarning;
1718 ctxt->vctxt.nodeMax = 0;
1719 ctxt->options |= XML_PARSE_DTDVALID;
1720 }
1721 ctxt->replaceEntities = xmlSubstituteEntitiesDefaultValue;
1722 if (ctxt->replaceEntities) {
1723 ctxt->options |= XML_PARSE_NOENT;
1724 }
1725 ctxt->record_info = 0;
1726 ctxt->nbChars = 0;
1727 ctxt->checkIndex = 0;
1728 ctxt->inSubset = 0;
1729 ctxt->errNo = XML_ERR_OK;
1730 ctxt->depth = 0;
1731 ctxt->charset = XML_CHAR_ENCODING_UTF8;
1732 ctxt->catalogs = NULL;
1733 ctxt->nbentities = 0;
1734 ctxt->sizeentities = 0;
1735 ctxt->sizeentcopy = 0;
1736 ctxt->input_id = 1;
1737 xmlInitNodeInfoSeq(&ctxt->node_seq);
1738 return(0);
1739 }
1740
1741 /**
1742 * xmlFreeParserCtxt:
1743 * @ctxt: an XML parser context
1744 *
1745 * Free all the memory used by a parser context. However the parsed
1746 * document in ctxt->myDoc is not freed.
1747 */
1748
1749 void
xmlFreeParserCtxt(xmlParserCtxtPtr ctxt)1750 xmlFreeParserCtxt(xmlParserCtxtPtr ctxt)
1751 {
1752 xmlParserInputPtr input;
1753
1754 if (ctxt == NULL) return;
1755
1756 while ((input = inputPop(ctxt)) != NULL) { /* Non consuming */
1757 xmlFreeInputStream(input);
1758 }
1759 if (ctxt->spaceTab != NULL) xmlFree(ctxt->spaceTab);
1760 if (ctxt->nameTab != NULL) xmlFree((xmlChar * *)ctxt->nameTab);
1761 if (ctxt->nodeTab != NULL) xmlFree(ctxt->nodeTab);
1762 if (ctxt->nodeInfoTab != NULL) xmlFree(ctxt->nodeInfoTab);
1763 if (ctxt->inputTab != NULL) xmlFree(ctxt->inputTab);
1764 if (ctxt->version != NULL) xmlFree((char *) ctxt->version);
1765 if (ctxt->encoding != NULL) xmlFree((char *) ctxt->encoding);
1766 if (ctxt->extSubURI != NULL) xmlFree((char *) ctxt->extSubURI);
1767 if (ctxt->extSubSystem != NULL) xmlFree((char *) ctxt->extSubSystem);
1768 #ifdef LIBXML_SAX1_ENABLED
1769 if ((ctxt->sax != NULL) &&
1770 (ctxt->sax != (xmlSAXHandlerPtr) &xmlDefaultSAXHandler))
1771 #else
1772 if (ctxt->sax != NULL)
1773 #endif /* LIBXML_SAX1_ENABLED */
1774 xmlFree(ctxt->sax);
1775 if (ctxt->directory != NULL) xmlFree((char *) ctxt->directory);
1776 if (ctxt->vctxt.nodeTab != NULL) xmlFree(ctxt->vctxt.nodeTab);
1777 if (ctxt->atts != NULL) xmlFree((xmlChar * *)ctxt->atts);
1778 if (ctxt->dict != NULL) xmlDictFree(ctxt->dict);
1779 if (ctxt->nsTab != NULL) xmlFree((char *) ctxt->nsTab);
1780 if (ctxt->pushTab != NULL) xmlFree(ctxt->pushTab);
1781 if (ctxt->attallocs != NULL) xmlFree(ctxt->attallocs);
1782 if (ctxt->attsDefault != NULL)
1783 xmlHashFree(ctxt->attsDefault, (xmlHashDeallocator) xmlFree);
1784 if (ctxt->attsSpecial != NULL)
1785 xmlHashFree(ctxt->attsSpecial, NULL);
1786 if (ctxt->freeElems != NULL) {
1787 xmlNodePtr cur, next;
1788
1789 cur = ctxt->freeElems;
1790 while (cur != NULL) {
1791 next = cur->next;
1792 xmlFree(cur);
1793 cur = next;
1794 }
1795 }
1796 if (ctxt->freeAttrs != NULL) {
1797 xmlAttrPtr cur, next;
1798
1799 cur = ctxt->freeAttrs;
1800 while (cur != NULL) {
1801 next = cur->next;
1802 xmlFree(cur);
1803 cur = next;
1804 }
1805 }
1806 /*
1807 * cleanup the error strings
1808 */
1809 if (ctxt->lastError.message != NULL)
1810 xmlFree(ctxt->lastError.message);
1811 if (ctxt->lastError.file != NULL)
1812 xmlFree(ctxt->lastError.file);
1813 if (ctxt->lastError.str1 != NULL)
1814 xmlFree(ctxt->lastError.str1);
1815 if (ctxt->lastError.str2 != NULL)
1816 xmlFree(ctxt->lastError.str2);
1817 if (ctxt->lastError.str3 != NULL)
1818 xmlFree(ctxt->lastError.str3);
1819
1820 #ifdef LIBXML_CATALOG_ENABLED
1821 if (ctxt->catalogs != NULL)
1822 xmlCatalogFreeLocal(ctxt->catalogs);
1823 #endif
1824 xmlFree(ctxt);
1825 }
1826
1827 /**
1828 * xmlNewParserCtxt:
1829 *
1830 * Allocate and initialize a new parser context.
1831 *
1832 * Returns the xmlParserCtxtPtr or NULL
1833 */
1834
1835 xmlParserCtxtPtr
xmlNewParserCtxt(void)1836 xmlNewParserCtxt(void)
1837 {
1838 xmlParserCtxtPtr ctxt;
1839
1840 ctxt = (xmlParserCtxtPtr) xmlMalloc(sizeof(xmlParserCtxt));
1841 if (ctxt == NULL) {
1842 xmlErrMemory(NULL, "cannot allocate parser context\n");
1843 return(NULL);
1844 }
1845 memset(ctxt, 0, sizeof(xmlParserCtxt));
1846 if (xmlInitParserCtxt(ctxt) < 0) {
1847 xmlFreeParserCtxt(ctxt);
1848 return(NULL);
1849 }
1850 return(ctxt);
1851 }
1852
1853 /************************************************************************
1854 * *
1855 * Handling of node informations *
1856 * *
1857 ************************************************************************/
1858
1859 /**
1860 * xmlClearParserCtxt:
1861 * @ctxt: an XML parser context
1862 *
1863 * Clear (release owned resources) and reinitialize a parser context
1864 */
1865
1866 void
xmlClearParserCtxt(xmlParserCtxtPtr ctxt)1867 xmlClearParserCtxt(xmlParserCtxtPtr ctxt)
1868 {
1869 if (ctxt==NULL)
1870 return;
1871 xmlClearNodeInfoSeq(&ctxt->node_seq);
1872 xmlCtxtReset(ctxt);
1873 }
1874
1875
1876 /**
1877 * xmlParserFindNodeInfo:
1878 * @ctx: an XML parser context
1879 * @node: an XML node within the tree
1880 *
1881 * Find the parser node info struct for a given node
1882 *
1883 * Returns an xmlParserNodeInfo block pointer or NULL
1884 */
1885 const xmlParserNodeInfo *
xmlParserFindNodeInfo(const xmlParserCtxtPtr ctx,const xmlNodePtr node)1886 xmlParserFindNodeInfo(const xmlParserCtxtPtr ctx, const xmlNodePtr node)
1887 {
1888 unsigned long pos;
1889
1890 if ((ctx == NULL) || (node == NULL))
1891 return (NULL);
1892 /* Find position where node should be at */
1893 pos = xmlParserFindNodeInfoIndex(&ctx->node_seq, node);
1894 if (pos < ctx->node_seq.length
1895 && ctx->node_seq.buffer[pos].node == node)
1896 return &ctx->node_seq.buffer[pos];
1897 else
1898 return NULL;
1899 }
1900
1901
1902 /**
1903 * xmlInitNodeInfoSeq:
1904 * @seq: a node info sequence pointer
1905 *
1906 * -- Initialize (set to initial state) node info sequence
1907 */
1908 void
xmlInitNodeInfoSeq(xmlParserNodeInfoSeqPtr seq)1909 xmlInitNodeInfoSeq(xmlParserNodeInfoSeqPtr seq)
1910 {
1911 if (seq == NULL)
1912 return;
1913 seq->length = 0;
1914 seq->maximum = 0;
1915 seq->buffer = NULL;
1916 }
1917
1918 /**
1919 * xmlClearNodeInfoSeq:
1920 * @seq: a node info sequence pointer
1921 *
1922 * -- Clear (release memory and reinitialize) node
1923 * info sequence
1924 */
1925 void
xmlClearNodeInfoSeq(xmlParserNodeInfoSeqPtr seq)1926 xmlClearNodeInfoSeq(xmlParserNodeInfoSeqPtr seq)
1927 {
1928 if (seq == NULL)
1929 return;
1930 if (seq->buffer != NULL)
1931 xmlFree(seq->buffer);
1932 xmlInitNodeInfoSeq(seq);
1933 }
1934
1935 /**
1936 * xmlParserFindNodeInfoIndex:
1937 * @seq: a node info sequence pointer
1938 * @node: an XML node pointer
1939 *
1940 *
1941 * xmlParserFindNodeInfoIndex : Find the index that the info record for
1942 * the given node is or should be at in a sorted sequence
1943 *
1944 * Returns a long indicating the position of the record
1945 */
1946 unsigned long
xmlParserFindNodeInfoIndex(const xmlParserNodeInfoSeqPtr seq,const xmlNodePtr node)1947 xmlParserFindNodeInfoIndex(const xmlParserNodeInfoSeqPtr seq,
1948 const xmlNodePtr node)
1949 {
1950 unsigned long upper, lower, middle;
1951 int found = 0;
1952
1953 if ((seq == NULL) || (node == NULL))
1954 return ((unsigned long) -1);
1955
1956 /* Do a binary search for the key */
1957 lower = 1;
1958 upper = seq->length;
1959 middle = 0;
1960 while (lower <= upper && !found) {
1961 middle = lower + (upper - lower) / 2;
1962 if (node == seq->buffer[middle - 1].node)
1963 found = 1;
1964 else if (node < seq->buffer[middle - 1].node)
1965 upper = middle - 1;
1966 else
1967 lower = middle + 1;
1968 }
1969
1970 /* Return position */
1971 if (middle == 0 || seq->buffer[middle - 1].node < node)
1972 return middle;
1973 else
1974 return middle - 1;
1975 }
1976
1977
1978 /**
1979 * xmlParserAddNodeInfo:
1980 * @ctxt: an XML parser context
1981 * @info: a node info sequence pointer
1982 *
1983 * Insert node info record into the sorted sequence
1984 */
1985 void
xmlParserAddNodeInfo(xmlParserCtxtPtr ctxt,const xmlParserNodeInfoPtr info)1986 xmlParserAddNodeInfo(xmlParserCtxtPtr ctxt,
1987 const xmlParserNodeInfoPtr info)
1988 {
1989 unsigned long pos;
1990
1991 if ((ctxt == NULL) || (info == NULL)) return;
1992
1993 /* Find pos and check to see if node is already in the sequence */
1994 pos = xmlParserFindNodeInfoIndex(&ctxt->node_seq, (xmlNodePtr)
1995 info->node);
1996
1997 if ((pos < ctxt->node_seq.length) &&
1998 (ctxt->node_seq.buffer != NULL) &&
1999 (ctxt->node_seq.buffer[pos].node == info->node)) {
2000 ctxt->node_seq.buffer[pos] = *info;
2001 }
2002
2003 /* Otherwise, we need to add new node to buffer */
2004 else {
2005 if ((ctxt->node_seq.length + 1 > ctxt->node_seq.maximum) ||
2006 (ctxt->node_seq.buffer == NULL)) {
2007 xmlParserNodeInfo *tmp_buffer;
2008 unsigned int byte_size;
2009
2010 if (ctxt->node_seq.maximum == 0)
2011 ctxt->node_seq.maximum = 2;
2012 byte_size = (sizeof(*ctxt->node_seq.buffer) *
2013 (2 * ctxt->node_seq.maximum));
2014
2015 if (ctxt->node_seq.buffer == NULL)
2016 tmp_buffer = (xmlParserNodeInfo *) xmlMalloc(byte_size);
2017 else
2018 tmp_buffer =
2019 (xmlParserNodeInfo *) xmlRealloc(ctxt->node_seq.buffer,
2020 byte_size);
2021
2022 if (tmp_buffer == NULL) {
2023 xmlErrMemory(ctxt, "failed to allocate buffer\n");
2024 return;
2025 }
2026 ctxt->node_seq.buffer = tmp_buffer;
2027 ctxt->node_seq.maximum *= 2;
2028 }
2029
2030 /* If position is not at end, move elements out of the way */
2031 if (pos != ctxt->node_seq.length) {
2032 unsigned long i;
2033
2034 for (i = ctxt->node_seq.length; i > pos; i--)
2035 ctxt->node_seq.buffer[i] = ctxt->node_seq.buffer[i - 1];
2036 }
2037
2038 /* Copy element and increase length */
2039 ctxt->node_seq.buffer[pos] = *info;
2040 ctxt->node_seq.length++;
2041 }
2042 }
2043
2044 /************************************************************************
2045 * *
2046 * Defaults settings *
2047 * *
2048 ************************************************************************/
2049 /**
2050 * xmlPedanticParserDefault:
2051 * @val: int 0 or 1
2052 *
2053 * Set and return the previous value for enabling pedantic warnings.
2054 *
2055 * Returns the last value for 0 for no substitution, 1 for substitution.
2056 */
2057
2058 int
xmlPedanticParserDefault(int val)2059 xmlPedanticParserDefault(int val) {
2060 int old = xmlPedanticParserDefaultValue;
2061
2062 xmlPedanticParserDefaultValue = val;
2063 return(old);
2064 }
2065
2066 /**
2067 * xmlLineNumbersDefault:
2068 * @val: int 0 or 1
2069 *
2070 * Set and return the previous value for enabling line numbers in elements
2071 * contents. This may break on old application and is turned off by default.
2072 *
2073 * Returns the last value for 0 for no substitution, 1 for substitution.
2074 */
2075
2076 int
xmlLineNumbersDefault(int val)2077 xmlLineNumbersDefault(int val) {
2078 int old = xmlLineNumbersDefaultValue;
2079
2080 xmlLineNumbersDefaultValue = val;
2081 return(old);
2082 }
2083
2084 /**
2085 * xmlSubstituteEntitiesDefault:
2086 * @val: int 0 or 1
2087 *
2088 * Set and return the previous value for default entity support.
2089 * Initially the parser always keep entity references instead of substituting
2090 * entity values in the output. This function has to be used to change the
2091 * default parser behavior
2092 * SAX::substituteEntities() has to be used for changing that on a file by
2093 * file basis.
2094 *
2095 * Returns the last value for 0 for no substitution, 1 for substitution.
2096 */
2097
2098 int
xmlSubstituteEntitiesDefault(int val)2099 xmlSubstituteEntitiesDefault(int val) {
2100 int old = xmlSubstituteEntitiesDefaultValue;
2101
2102 xmlSubstituteEntitiesDefaultValue = val;
2103 return(old);
2104 }
2105
2106 /**
2107 * xmlKeepBlanksDefault:
2108 * @val: int 0 or 1
2109 *
2110 * Set and return the previous value for default blanks text nodes support.
2111 * The 1.x version of the parser used an heuristic to try to detect
2112 * ignorable white spaces. As a result the SAX callback was generating
2113 * xmlSAX2IgnorableWhitespace() callbacks instead of characters() one, and when
2114 * using the DOM output text nodes containing those blanks were not generated.
2115 * The 2.x and later version will switch to the XML standard way and
2116 * ignorableWhitespace() are only generated when running the parser in
2117 * validating mode and when the current element doesn't allow CDATA or
2118 * mixed content.
2119 * This function is provided as a way to force the standard behavior
2120 * on 1.X libs and to switch back to the old mode for compatibility when
2121 * running 1.X client code on 2.X . Upgrade of 1.X code should be done
2122 * by using xmlIsBlankNode() commodity function to detect the "empty"
2123 * nodes generated.
2124 * This value also affect autogeneration of indentation when saving code
2125 * if blanks sections are kept, indentation is not generated.
2126 *
2127 * Returns the last value for 0 for no substitution, 1 for substitution.
2128 */
2129
2130 int
xmlKeepBlanksDefault(int val)2131 xmlKeepBlanksDefault(int val) {
2132 int old = xmlKeepBlanksDefaultValue;
2133
2134 xmlKeepBlanksDefaultValue = val;
2135 if (!val) xmlIndentTreeOutput = 1;
2136 return(old);
2137 }
2138
2139 #define bottom_parserInternals
2140 #include "elfgcchack.h"
2141