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 LIBXML_ZLIB_ENABLED
36 #include <vtk_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 #define CUR(ctxt) ctxt->input->cur
59 #define END(ctxt) ctxt->input->end
60 #define VALID_CTXT(ctxt) (CUR(ctxt) <= END(ctxt))
61
62 #include "buf.h"
63 #include "enc.h"
64
65 /*
66 * Various global defaults for parsing
67 */
68
69 /**
70 * xmlCheckVersion:
71 * @version: the include version number
72 *
73 * check the compiled lib version against the include one.
74 * This can warn or immediately kill the application
75 */
76 void
xmlCheckVersion(int version)77 xmlCheckVersion(int version) {
78 int myversion = (int) LIBXML_VERSION;
79
80 xmlInitParser();
81
82 if ((myversion / 10000) != (version / 10000)) {
83 xmlGenericError(xmlGenericErrorContext,
84 "Fatal: program compiled against libxml %d using libxml %d\n",
85 (version / 10000), (myversion / 10000));
86 fprintf(stderr,
87 "Fatal: program compiled against libxml %d using libxml %d\n",
88 (version / 10000), (myversion / 10000));
89 }
90 if ((myversion / 100) < (version / 100)) {
91 xmlGenericError(xmlGenericErrorContext,
92 "Warning: program compiled against libxml %d using older %d\n",
93 (version / 100), (myversion / 100));
94 }
95 }
96
97
98 /************************************************************************
99 * *
100 * Some factorized error routines *
101 * *
102 ************************************************************************/
103
104
105 /**
106 * xmlErrMemory:
107 * @ctxt: an XML parser context
108 * @extra: extra information
109 *
110 * Handle a redefinition of attribute error
111 */
112 void
xmlErrMemory(xmlParserCtxtPtr ctxt,const char * extra)113 xmlErrMemory(xmlParserCtxtPtr ctxt, const char *extra)
114 {
115 if ((ctxt != NULL) && (ctxt->disableSAX != 0) &&
116 (ctxt->instate == XML_PARSER_EOF))
117 return;
118 if (ctxt != NULL) {
119 ctxt->errNo = XML_ERR_NO_MEMORY;
120 ctxt->instate = XML_PARSER_EOF;
121 ctxt->disableSAX = 1;
122 }
123 if (extra)
124 __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER,
125 XML_ERR_NO_MEMORY, XML_ERR_FATAL, NULL, 0, extra,
126 NULL, NULL, 0, 0,
127 "Memory allocation failed : %s\n", extra);
128 else
129 __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER,
130 XML_ERR_NO_MEMORY, XML_ERR_FATAL, NULL, 0, NULL,
131 NULL, NULL, 0, 0, "Memory allocation failed\n");
132 }
133
134 /**
135 * __xmlErrEncoding:
136 * @ctxt: an XML parser context
137 * @xmlerr: the error number
138 * @msg: the error message
139 * @str1: an string info
140 * @str2: an string info
141 *
142 * Handle an encoding error
143 */
144 void
__xmlErrEncoding(xmlParserCtxtPtr ctxt,xmlParserErrors xmlerr,const char * msg,const xmlChar * str1,const xmlChar * str2)145 __xmlErrEncoding(xmlParserCtxtPtr ctxt, xmlParserErrors xmlerr,
146 const char *msg, const xmlChar * str1, const xmlChar * str2)
147 {
148 if ((ctxt != NULL) && (ctxt->disableSAX != 0) &&
149 (ctxt->instate == XML_PARSER_EOF))
150 return;
151 if (ctxt != NULL)
152 ctxt->errNo = xmlerr;
153 __xmlRaiseError(NULL, NULL, NULL,
154 ctxt, NULL, XML_FROM_PARSER, xmlerr, XML_ERR_FATAL,
155 NULL, 0, (const char *) str1, (const char *) str2,
156 NULL, 0, 0, msg, str1, str2);
157 if (ctxt != NULL) {
158 ctxt->wellFormed = 0;
159 if (ctxt->recovery == 0)
160 ctxt->disableSAX = 1;
161 }
162 }
163
164 /**
165 * xmlErrInternal:
166 * @ctxt: an XML parser context
167 * @msg: the error message
168 * @str: error information
169 *
170 * Handle an internal error
171 */
172 static void LIBXML_ATTR_FORMAT(2,0)
xmlErrInternal(xmlParserCtxtPtr ctxt,const char * msg,const xmlChar * str)173 xmlErrInternal(xmlParserCtxtPtr ctxt, const char *msg, const xmlChar * str)
174 {
175 if ((ctxt != NULL) && (ctxt->disableSAX != 0) &&
176 (ctxt->instate == XML_PARSER_EOF))
177 return;
178 if (ctxt != NULL)
179 ctxt->errNo = XML_ERR_INTERNAL_ERROR;
180 __xmlRaiseError(NULL, NULL, NULL,
181 ctxt, NULL, XML_FROM_PARSER, XML_ERR_INTERNAL_ERROR,
182 XML_ERR_FATAL, NULL, 0, (const char *) str, NULL, NULL,
183 0, 0, msg, str);
184 if (ctxt != NULL) {
185 ctxt->wellFormed = 0;
186 if (ctxt->recovery == 0)
187 ctxt->disableSAX = 1;
188 }
189 }
190
191 /**
192 * xmlErrEncodingInt:
193 * @ctxt: an XML parser context
194 * @error: the error number
195 * @msg: the error message
196 * @val: an integer value
197 *
198 * n encoding error
199 */
200 static void LIBXML_ATTR_FORMAT(3,0)
xmlErrEncodingInt(xmlParserCtxtPtr ctxt,xmlParserErrors error,const char * msg,int val)201 xmlErrEncodingInt(xmlParserCtxtPtr ctxt, xmlParserErrors error,
202 const char *msg, int val)
203 {
204 if ((ctxt != NULL) && (ctxt->disableSAX != 0) &&
205 (ctxt->instate == XML_PARSER_EOF))
206 return;
207 if (ctxt != NULL)
208 ctxt->errNo = error;
209 __xmlRaiseError(NULL, NULL, NULL,
210 ctxt, NULL, XML_FROM_PARSER, error, XML_ERR_FATAL,
211 NULL, 0, NULL, NULL, NULL, val, 0, msg, val);
212 if (ctxt != NULL) {
213 ctxt->wellFormed = 0;
214 if (ctxt->recovery == 0)
215 ctxt->disableSAX = 1;
216 }
217 }
218
219 /**
220 * xmlIsLetter:
221 * @c: an unicode character (int)
222 *
223 * Check whether the character is allowed by the production
224 * [84] Letter ::= BaseChar | Ideographic
225 *
226 * Returns 0 if not, non-zero otherwise
227 */
228 int
xmlIsLetter(int c)229 xmlIsLetter(int c) {
230 return(IS_BASECHAR(c) || IS_IDEOGRAPHIC(c));
231 }
232
233 /************************************************************************
234 * *
235 * Input handling functions for progressive parsing *
236 * *
237 ************************************************************************/
238
239 /* #define DEBUG_INPUT */
240 /* #define DEBUG_STACK */
241 /* #define DEBUG_PUSH */
242
243
244 /* we need to keep enough input to show errors in context */
245 #define LINE_LEN 80
246
247 #ifdef DEBUG_INPUT
248 #define CHECK_BUFFER(in) check_buffer(in)
249
250 static
check_buffer(xmlParserInputPtr in)251 void check_buffer(xmlParserInputPtr in) {
252 if (in->base != xmlBufContent(in->buf->buffer)) {
253 xmlGenericError(xmlGenericErrorContext,
254 "xmlParserInput: base mismatch problem\n");
255 }
256 if (in->cur < in->base) {
257 xmlGenericError(xmlGenericErrorContext,
258 "xmlParserInput: cur < base problem\n");
259 }
260 if (in->cur > in->base + xmlBufUse(in->buf->buffer)) {
261 xmlGenericError(xmlGenericErrorContext,
262 "xmlParserInput: cur > base + use problem\n");
263 }
264 xmlGenericError(xmlGenericErrorContext,"buffer %x : content %x, cur %d, use %d\n",
265 (int) in, (int) xmlBufContent(in->buf->buffer), in->cur - in->base,
266 xmlBufUse(in->buf->buffer));
267 }
268
269 #else
270 #define CHECK_BUFFER(in)
271 #endif
272
273
274 /**
275 * xmlParserInputRead:
276 * @in: an XML parser input
277 * @len: an indicative size for the lookahead
278 *
279 * This function was internal and is deprecated.
280 *
281 * Returns -1 as this is an error to use it.
282 */
283 int
xmlParserInputRead(xmlParserInputPtr in ATTRIBUTE_UNUSED,int len ATTRIBUTE_UNUSED)284 xmlParserInputRead(xmlParserInputPtr in ATTRIBUTE_UNUSED, int len ATTRIBUTE_UNUSED) {
285 return(-1);
286 }
287
288 /**
289 * xmlParserInputGrow:
290 * @in: an XML parser input
291 * @len: an indicative size for the lookahead
292 *
293 * This function increase the input for the parser. It tries to
294 * preserve pointers to the input buffer, and keep already read data
295 *
296 * Returns the amount of char read, or -1 in case of error, 0 indicate the
297 * end of this entity
298 */
299 int
xmlParserInputGrow(xmlParserInputPtr in,int len)300 xmlParserInputGrow(xmlParserInputPtr in, int len) {
301 int ret;
302 size_t indx;
303 const xmlChar *content;
304
305 if ((in == NULL) || (len < 0)) return(-1);
306 #ifdef DEBUG_INPUT
307 xmlGenericError(xmlGenericErrorContext, "Grow\n");
308 #endif
309 if (in->buf == NULL) return(-1);
310 if (in->base == NULL) return(-1);
311 if (in->cur == NULL) return(-1);
312 if (in->buf->buffer == NULL) return(-1);
313
314 CHECK_BUFFER(in);
315
316 indx = in->cur - in->base;
317 if (xmlBufUse(in->buf->buffer) > (unsigned int) indx + INPUT_CHUNK) {
318
319 CHECK_BUFFER(in);
320
321 return(0);
322 }
323 if (in->buf->readcallback != NULL) {
324 ret = xmlParserInputBufferGrow(in->buf, len);
325 } else
326 return(0);
327
328 /*
329 * NOTE : in->base may be a "dangling" i.e. freed pointer in this
330 * block, but we use it really as an integer to do some
331 * pointer arithmetic. Insure will raise it as a bug but in
332 * that specific case, that's not !
333 */
334
335 content = xmlBufContent(in->buf->buffer);
336 if (in->base != content) {
337 /*
338 * the buffer has been reallocated
339 */
340 indx = in->cur - in->base;
341 in->base = content;
342 in->cur = &content[indx];
343 }
344 in->end = xmlBufEnd(in->buf->buffer);
345
346 CHECK_BUFFER(in);
347
348 return(ret);
349 }
350
351 /**
352 * xmlParserInputShrink:
353 * @in: an XML parser input
354 *
355 * This function removes used input for the parser.
356 */
357 void
xmlParserInputShrink(xmlParserInputPtr in)358 xmlParserInputShrink(xmlParserInputPtr in) {
359 size_t used;
360 size_t ret;
361 size_t indx;
362 const xmlChar *content;
363
364 #ifdef DEBUG_INPUT
365 xmlGenericError(xmlGenericErrorContext, "Shrink\n");
366 #endif
367 if (in == NULL) return;
368 if (in->buf == NULL) return;
369 if (in->base == NULL) return;
370 if (in->cur == NULL) return;
371 if (in->buf->buffer == NULL) return;
372
373 CHECK_BUFFER(in);
374
375 used = in->cur - xmlBufContent(in->buf->buffer);
376 /*
377 * Do not shrink on large buffers whose only a tiny fraction
378 * was consumed
379 */
380 if (used > INPUT_CHUNK) {
381 ret = xmlBufShrink(in->buf->buffer, used - LINE_LEN);
382 if (ret > 0) {
383 in->cur -= ret;
384 in->consumed += ret;
385 }
386 in->end = xmlBufEnd(in->buf->buffer);
387 }
388
389 CHECK_BUFFER(in);
390
391 if (xmlBufUse(in->buf->buffer) > INPUT_CHUNK) {
392 return;
393 }
394 xmlParserInputBufferRead(in->buf, 2 * INPUT_CHUNK);
395 content = xmlBufContent(in->buf->buffer);
396 if (in->base != content) {
397 /*
398 * the buffer has been reallocated
399 */
400 indx = in->cur - in->base;
401 in->base = content;
402 in->cur = &content[indx];
403 }
404 in->end = xmlBufEnd(in->buf->buffer);
405
406 CHECK_BUFFER(in);
407 }
408
409 /************************************************************************
410 * *
411 * UTF8 character input and related functions *
412 * *
413 ************************************************************************/
414
415 /**
416 * xmlNextChar:
417 * @ctxt: the XML parser context
418 *
419 * Skip to the next char input char.
420 */
421
422 void
xmlNextChar(xmlParserCtxtPtr ctxt)423 xmlNextChar(xmlParserCtxtPtr ctxt)
424 {
425 if ((ctxt == NULL) || (ctxt->instate == XML_PARSER_EOF) ||
426 (ctxt->input == NULL))
427 return;
428
429 if (!(VALID_CTXT(ctxt))) {
430 xmlErrInternal(ctxt, "Parser input data memory error\n", NULL);
431 ctxt->errNo = XML_ERR_INTERNAL_ERROR;
432 xmlStopParser(ctxt);
433 return;
434 }
435
436 if ((*ctxt->input->cur == 0) &&
437 (xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0)) {
438 return;
439 }
440
441 if (ctxt->charset == XML_CHAR_ENCODING_UTF8) {
442 const unsigned char *cur;
443 unsigned char c;
444
445 /*
446 * 2.11 End-of-Line Handling
447 * the literal two-character sequence "#xD#xA" or a standalone
448 * literal #xD, an XML processor must pass to the application
449 * the single character #xA.
450 */
451 if (*(ctxt->input->cur) == '\n') {
452 ctxt->input->line++; ctxt->input->col = 1;
453 } else
454 ctxt->input->col++;
455
456 /*
457 * We are supposed to handle UTF8, check it's valid
458 * From rfc2044: encoding of the Unicode values on UTF-8:
459 *
460 * UCS-4 range (hex.) UTF-8 octet sequence (binary)
461 * 0000 0000-0000 007F 0xxxxxxx
462 * 0000 0080-0000 07FF 110xxxxx 10xxxxxx
463 * 0000 0800-0000 FFFF 1110xxxx 10xxxxxx 10xxxxxx
464 *
465 * Check for the 0x110000 limit too
466 */
467 cur = ctxt->input->cur;
468
469 c = *cur;
470 if (c & 0x80) {
471 if (c == 0xC0)
472 goto encoding_error;
473 if (cur[1] == 0) {
474 xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
475 cur = ctxt->input->cur;
476 }
477 if ((cur[1] & 0xc0) != 0x80)
478 goto encoding_error;
479 if ((c & 0xe0) == 0xe0) {
480 unsigned int val;
481
482 if (cur[2] == 0) {
483 xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
484 cur = ctxt->input->cur;
485 }
486 if ((cur[2] & 0xc0) != 0x80)
487 goto encoding_error;
488 if ((c & 0xf0) == 0xf0) {
489 if (cur[3] == 0) {
490 xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
491 cur = ctxt->input->cur;
492 }
493 if (((c & 0xf8) != 0xf0) ||
494 ((cur[3] & 0xc0) != 0x80))
495 goto encoding_error;
496 /* 4-byte code */
497 ctxt->input->cur += 4;
498 val = (cur[0] & 0x7) << 18;
499 val |= (cur[1] & 0x3f) << 12;
500 val |= (cur[2] & 0x3f) << 6;
501 val |= cur[3] & 0x3f;
502 } else {
503 /* 3-byte code */
504 ctxt->input->cur += 3;
505 val = (cur[0] & 0xf) << 12;
506 val |= (cur[1] & 0x3f) << 6;
507 val |= cur[2] & 0x3f;
508 }
509 if (((val > 0xd7ff) && (val < 0xe000)) ||
510 ((val > 0xfffd) && (val < 0x10000)) ||
511 (val >= 0x110000)) {
512 xmlErrEncodingInt(ctxt, XML_ERR_INVALID_CHAR,
513 "Char 0x%X out of allowed range\n",
514 val);
515 }
516 } else
517 /* 2-byte code */
518 ctxt->input->cur += 2;
519 } else
520 /* 1-byte code */
521 ctxt->input->cur++;
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 }
535 if (*ctxt->input->cur == 0)
536 xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
537 return;
538 encoding_error:
539 /*
540 * If we detect an UTF8 error that probably mean that the
541 * input encoding didn't get properly advertised in the
542 * declaration header. Report the error and switch the encoding
543 * to ISO-Latin-1 (if you don't like this policy, just declare the
544 * encoding !)
545 */
546 if ((ctxt == NULL) || (ctxt->input == NULL) ||
547 (ctxt->input->end - ctxt->input->cur < 4)) {
548 __xmlErrEncoding(ctxt, XML_ERR_INVALID_CHAR,
549 "Input is not proper UTF-8, indicate encoding !\n",
550 NULL, NULL);
551 } else {
552 char buffer[150];
553
554 snprintf(buffer, 149, "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n",
555 ctxt->input->cur[0], ctxt->input->cur[1],
556 ctxt->input->cur[2], ctxt->input->cur[3]);
557 __xmlErrEncoding(ctxt, XML_ERR_INVALID_CHAR,
558 "Input is not proper UTF-8, indicate encoding !\n%s",
559 BAD_CAST buffer, NULL);
560 }
561 ctxt->charset = XML_CHAR_ENCODING_8859_1;
562 ctxt->input->cur++;
563 return;
564 }
565
566 /**
567 * xmlCurrentChar:
568 * @ctxt: the XML parser context
569 * @len: pointer to the length of the char read
570 *
571 * The current char value, if using UTF-8 this may actually span multiple
572 * bytes in the input buffer. Implement the end of line normalization:
573 * 2.11 End-of-Line Handling
574 * Wherever an external parsed entity or the literal entity value
575 * of an internal parsed entity contains either the literal two-character
576 * sequence "#xD#xA" or a standalone literal #xD, an XML processor
577 * must pass to the application the single character #xA.
578 * This behavior can conveniently be produced by normalizing all
579 * line breaks to #xA on input, before parsing.)
580 *
581 * Returns the current char value and its length
582 */
583
584 int
xmlCurrentChar(xmlParserCtxtPtr ctxt,int * len)585 xmlCurrentChar(xmlParserCtxtPtr ctxt, int *len) {
586 if ((ctxt == NULL) || (len == NULL) || (ctxt->input == NULL)) return(0);
587 if (ctxt->instate == XML_PARSER_EOF)
588 return(0);
589
590 if ((*ctxt->input->cur >= 0x20) && (*ctxt->input->cur <= 0x7F)) {
591 *len = 1;
592 return((int) *ctxt->input->cur);
593 }
594 if (ctxt->charset == XML_CHAR_ENCODING_UTF8) {
595 /*
596 * We are supposed to handle UTF8, check it's valid
597 * From rfc2044: encoding of the Unicode values on UTF-8:
598 *
599 * UCS-4 range (hex.) UTF-8 octet sequence (binary)
600 * 0000 0000-0000 007F 0xxxxxxx
601 * 0000 0080-0000 07FF 110xxxxx 10xxxxxx
602 * 0000 0800-0000 FFFF 1110xxxx 10xxxxxx 10xxxxxx
603 *
604 * Check for the 0x110000 limit too
605 */
606 const unsigned char *cur = ctxt->input->cur;
607 unsigned char c;
608 unsigned int val;
609
610 c = *cur;
611 if (c & 0x80) {
612 if (((c & 0x40) == 0) || (c == 0xC0))
613 goto encoding_error;
614 if (cur[1] == 0) {
615 xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
616 cur = ctxt->input->cur;
617 }
618 if ((cur[1] & 0xc0) != 0x80)
619 goto encoding_error;
620 if ((c & 0xe0) == 0xe0) {
621 if (cur[2] == 0) {
622 xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
623 cur = ctxt->input->cur;
624 }
625 if ((cur[2] & 0xc0) != 0x80)
626 goto encoding_error;
627 if ((c & 0xf0) == 0xf0) {
628 if (cur[3] == 0) {
629 xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
630 cur = ctxt->input->cur;
631 }
632 if (((c & 0xf8) != 0xf0) ||
633 ((cur[3] & 0xc0) != 0x80))
634 goto encoding_error;
635 /* 4-byte code */
636 *len = 4;
637 val = (cur[0] & 0x7) << 18;
638 val |= (cur[1] & 0x3f) << 12;
639 val |= (cur[2] & 0x3f) << 6;
640 val |= cur[3] & 0x3f;
641 if (val < 0x10000)
642 goto encoding_error;
643 } else {
644 /* 3-byte code */
645 *len = 3;
646 val = (cur[0] & 0xf) << 12;
647 val |= (cur[1] & 0x3f) << 6;
648 val |= cur[2] & 0x3f;
649 if (val < 0x800)
650 goto encoding_error;
651 }
652 } else {
653 /* 2-byte code */
654 *len = 2;
655 val = (cur[0] & 0x1f) << 6;
656 val |= cur[1] & 0x3f;
657 if (val < 0x80)
658 goto encoding_error;
659 }
660 if (!IS_CHAR(val)) {
661 xmlErrEncodingInt(ctxt, XML_ERR_INVALID_CHAR,
662 "Char 0x%X out of allowed range\n", val);
663 }
664 return(val);
665 } else {
666 /* 1-byte code */
667 *len = 1;
668 if (*ctxt->input->cur == 0)
669 xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
670 if ((*ctxt->input->cur == 0) &&
671 (ctxt->input->end > ctxt->input->cur)) {
672 xmlErrEncodingInt(ctxt, XML_ERR_INVALID_CHAR,
673 "Char 0x0 out of allowed range\n", 0);
674 }
675 if (*ctxt->input->cur == 0xD) {
676 if (ctxt->input->cur[1] == 0xA) {
677 ctxt->input->cur++;
678 }
679 return(0xA);
680 }
681 return((int) *ctxt->input->cur);
682 }
683 }
684 /*
685 * Assume it's a fixed length encoding (1) with
686 * a compatible encoding for the ASCII set, since
687 * XML constructs only use < 128 chars
688 */
689 *len = 1;
690 if (*ctxt->input->cur == 0xD) {
691 if (ctxt->input->cur[1] == 0xA) {
692 ctxt->input->cur++;
693 }
694 return(0xA);
695 }
696 return((int) *ctxt->input->cur);
697 encoding_error:
698 /*
699 * An encoding problem may arise from a truncated input buffer
700 * splitting a character in the middle. In that case do not raise
701 * an error but return 0 to indicate an end of stream problem
702 */
703 if (ctxt->input->end - ctxt->input->cur < 4) {
704 *len = 0;
705 return(0);
706 }
707
708 /*
709 * If we detect an UTF8 error that probably mean that the
710 * input encoding didn't get properly advertised in the
711 * declaration header. Report the error and switch the encoding
712 * to ISO-Latin-1 (if you don't like this policy, just declare the
713 * encoding !)
714 */
715 {
716 char buffer[150];
717
718 snprintf(&buffer[0], 149, "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n",
719 ctxt->input->cur[0], ctxt->input->cur[1],
720 ctxt->input->cur[2], ctxt->input->cur[3]);
721 __xmlErrEncoding(ctxt, XML_ERR_INVALID_CHAR,
722 "Input is not proper UTF-8, indicate encoding !\n%s",
723 BAD_CAST buffer, NULL);
724 }
725 ctxt->charset = XML_CHAR_ENCODING_8859_1;
726 *len = 1;
727 return((int) *ctxt->input->cur);
728 }
729
730 /**
731 * xmlStringCurrentChar:
732 * @ctxt: the XML parser context
733 * @cur: pointer to the beginning of the char
734 * @len: pointer to the length of the char read
735 *
736 * The current char value, if using UTF-8 this may actually span multiple
737 * bytes in the input buffer.
738 *
739 * Returns the current char value and its length
740 */
741
742 int
xmlStringCurrentChar(xmlParserCtxtPtr ctxt,const xmlChar * cur,int * len)743 xmlStringCurrentChar(xmlParserCtxtPtr ctxt, const xmlChar * cur, int *len)
744 {
745 if ((len == NULL) || (cur == NULL)) return(0);
746 if ((ctxt == NULL) || (ctxt->charset == XML_CHAR_ENCODING_UTF8)) {
747 /*
748 * We are supposed to handle UTF8, check it's valid
749 * From rfc2044: encoding of the Unicode values on UTF-8:
750 *
751 * UCS-4 range (hex.) UTF-8 octet sequence (binary)
752 * 0000 0000-0000 007F 0xxxxxxx
753 * 0000 0080-0000 07FF 110xxxxx 10xxxxxx
754 * 0000 0800-0000 FFFF 1110xxxx 10xxxxxx 10xxxxxx
755 *
756 * Check for the 0x110000 limit too
757 */
758 unsigned char c;
759 unsigned int val;
760
761 c = *cur;
762 if (c & 0x80) {
763 if ((cur[1] & 0xc0) != 0x80)
764 goto encoding_error;
765 if ((c & 0xe0) == 0xe0) {
766
767 if ((cur[2] & 0xc0) != 0x80)
768 goto encoding_error;
769 if ((c & 0xf0) == 0xf0) {
770 if (((c & 0xf8) != 0xf0) || ((cur[3] & 0xc0) != 0x80))
771 goto encoding_error;
772 /* 4-byte code */
773 *len = 4;
774 val = (cur[0] & 0x7) << 18;
775 val |= (cur[1] & 0x3f) << 12;
776 val |= (cur[2] & 0x3f) << 6;
777 val |= cur[3] & 0x3f;
778 } else {
779 /* 3-byte code */
780 *len = 3;
781 val = (cur[0] & 0xf) << 12;
782 val |= (cur[1] & 0x3f) << 6;
783 val |= cur[2] & 0x3f;
784 }
785 } else {
786 /* 2-byte code */
787 *len = 2;
788 val = (cur[0] & 0x1f) << 6;
789 val |= cur[1] & 0x3f;
790 }
791 if (!IS_CHAR(val)) {
792 xmlErrEncodingInt(ctxt, XML_ERR_INVALID_CHAR,
793 "Char 0x%X out of allowed range\n", val);
794 }
795 return (val);
796 } else {
797 /* 1-byte code */
798 *len = 1;
799 return ((int) *cur);
800 }
801 }
802 /*
803 * Assume it's a fixed length encoding (1) with
804 * a compatible encoding for the ASCII set, since
805 * XML constructs only use < 128 chars
806 */
807 *len = 1;
808 return ((int) *cur);
809 encoding_error:
810
811 /*
812 * An encoding problem may arise from a truncated input buffer
813 * splitting a character in the middle. In that case do not raise
814 * an error but return 0 to indicate an end of stream problem
815 */
816 if ((ctxt == NULL) || (ctxt->input == NULL) ||
817 (ctxt->input->end - ctxt->input->cur < 4)) {
818 *len = 0;
819 return(0);
820 }
821 /*
822 * If we detect an UTF8 error that probably mean that the
823 * input encoding didn't get properly advertised in the
824 * declaration header. Report the error and switch the encoding
825 * to ISO-Latin-1 (if you don't like this policy, just declare the
826 * encoding !)
827 */
828 {
829 char buffer[150];
830
831 snprintf(buffer, 149, "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n",
832 ctxt->input->cur[0], ctxt->input->cur[1],
833 ctxt->input->cur[2], ctxt->input->cur[3]);
834 __xmlErrEncoding(ctxt, XML_ERR_INVALID_CHAR,
835 "Input is not proper UTF-8, indicate encoding !\n%s",
836 BAD_CAST buffer, NULL);
837 }
838 *len = 1;
839 return ((int) *cur);
840 }
841
842 /**
843 * xmlCopyCharMultiByte:
844 * @out: pointer to an array of xmlChar
845 * @val: the char value
846 *
847 * append the char value in the array
848 *
849 * Returns the number of xmlChar written
850 */
851 int
xmlCopyCharMultiByte(xmlChar * out,int val)852 xmlCopyCharMultiByte(xmlChar *out, int val) {
853 if (out == NULL) return(0);
854 /*
855 * We are supposed to handle UTF8, check it's valid
856 * From rfc2044: encoding of the Unicode values on UTF-8:
857 *
858 * UCS-4 range (hex.) UTF-8 octet sequence (binary)
859 * 0000 0000-0000 007F 0xxxxxxx
860 * 0000 0080-0000 07FF 110xxxxx 10xxxxxx
861 * 0000 0800-0000 FFFF 1110xxxx 10xxxxxx 10xxxxxx
862 */
863 if (val >= 0x80) {
864 xmlChar *savedout = out;
865 int bits;
866 if (val < 0x800) { *out++= (val >> 6) | 0xC0; bits= 0; }
867 else if (val < 0x10000) { *out++= (val >> 12) | 0xE0; bits= 6;}
868 else if (val < 0x110000) { *out++= (val >> 18) | 0xF0; bits= 12; }
869 else {
870 xmlErrEncodingInt(NULL, XML_ERR_INVALID_CHAR,
871 "Internal error, xmlCopyCharMultiByte 0x%X out of bound\n",
872 val);
873 return(0);
874 }
875 for ( ; bits >= 0; bits-= 6)
876 *out++= ((val >> bits) & 0x3F) | 0x80 ;
877 return (out - savedout);
878 }
879 *out = (xmlChar) val;
880 return 1;
881 }
882
883 /**
884 * xmlCopyChar:
885 * @len: Ignored, compatibility
886 * @out: pointer to an array of xmlChar
887 * @val: the char value
888 *
889 * append the char value in the array
890 *
891 * Returns the number of xmlChar written
892 */
893
894 int
xmlCopyChar(int len ATTRIBUTE_UNUSED,xmlChar * out,int val)895 xmlCopyChar(int len ATTRIBUTE_UNUSED, xmlChar *out, int val) {
896 if (out == NULL) return(0);
897 /* the len parameter is ignored */
898 if (val >= 0x80) {
899 return(xmlCopyCharMultiByte (out, val));
900 }
901 *out = (xmlChar) val;
902 return 1;
903 }
904
905 /************************************************************************
906 * *
907 * Commodity functions to switch encodings *
908 * *
909 ************************************************************************/
910
911 static int
912 xmlSwitchToEncodingInt(xmlParserCtxtPtr ctxt,
913 xmlCharEncodingHandlerPtr handler, int len);
914 static int
915 xmlSwitchInputEncodingInt(xmlParserCtxtPtr ctxt, xmlParserInputPtr input,
916 xmlCharEncodingHandlerPtr handler, int len);
917 /**
918 * xmlSwitchEncoding:
919 * @ctxt: the parser context
920 * @enc: the encoding value (number)
921 *
922 * change the input functions when discovering the character encoding
923 * of a given entity.
924 *
925 * Returns 0 in case of success, -1 otherwise
926 */
927 int
xmlSwitchEncoding(xmlParserCtxtPtr ctxt,xmlCharEncoding enc)928 xmlSwitchEncoding(xmlParserCtxtPtr ctxt, xmlCharEncoding enc)
929 {
930 xmlCharEncodingHandlerPtr handler;
931 int len = -1;
932 int ret;
933
934 if (ctxt == NULL) return(-1);
935 switch (enc) {
936 case XML_CHAR_ENCODING_ERROR:
937 __xmlErrEncoding(ctxt, XML_ERR_UNKNOWN_ENCODING,
938 "encoding unknown\n", NULL, NULL);
939 return(-1);
940 case XML_CHAR_ENCODING_NONE:
941 /* let's assume it's UTF-8 without the XML decl */
942 ctxt->charset = XML_CHAR_ENCODING_UTF8;
943 return(0);
944 case XML_CHAR_ENCODING_UTF8:
945 /* default encoding, no conversion should be needed */
946 ctxt->charset = XML_CHAR_ENCODING_UTF8;
947
948 /*
949 * Errata on XML-1.0 June 20 2001
950 * Specific handling of the Byte Order Mark for
951 * UTF-8
952 */
953 if ((ctxt->input != NULL) &&
954 (ctxt->input->cur[0] == 0xEF) &&
955 (ctxt->input->cur[1] == 0xBB) &&
956 (ctxt->input->cur[2] == 0xBF)) {
957 ctxt->input->cur += 3;
958 }
959 return(0);
960 case XML_CHAR_ENCODING_UTF16LE:
961 case XML_CHAR_ENCODING_UTF16BE:
962 /*The raw input characters are encoded
963 *in UTF-16. As we expect this function
964 *to be called after xmlCharEncInFunc, we expect
965 *ctxt->input->cur to contain UTF-8 encoded characters.
966 *So the raw UTF16 Byte Order Mark
967 *has also been converted into
968 *an UTF-8 BOM. Let's skip that BOM.
969 */
970 if ((ctxt->input != NULL) && (ctxt->input->cur != NULL) &&
971 (ctxt->input->cur[0] == 0xEF) &&
972 (ctxt->input->cur[1] == 0xBB) &&
973 (ctxt->input->cur[2] == 0xBF)) {
974 ctxt->input->cur += 3;
975 }
976 len = 90;
977 break;
978 case XML_CHAR_ENCODING_UCS2:
979 len = 90;
980 break;
981 case XML_CHAR_ENCODING_UCS4BE:
982 case XML_CHAR_ENCODING_UCS4LE:
983 case XML_CHAR_ENCODING_UCS4_2143:
984 case XML_CHAR_ENCODING_UCS4_3412:
985 len = 180;
986 break;
987 case XML_CHAR_ENCODING_EBCDIC:
988 case XML_CHAR_ENCODING_8859_1:
989 case XML_CHAR_ENCODING_8859_2:
990 case XML_CHAR_ENCODING_8859_3:
991 case XML_CHAR_ENCODING_8859_4:
992 case XML_CHAR_ENCODING_8859_5:
993 case XML_CHAR_ENCODING_8859_6:
994 case XML_CHAR_ENCODING_8859_7:
995 case XML_CHAR_ENCODING_8859_8:
996 case XML_CHAR_ENCODING_8859_9:
997 case XML_CHAR_ENCODING_ASCII:
998 case XML_CHAR_ENCODING_2022_JP:
999 case XML_CHAR_ENCODING_SHIFT_JIS:
1000 case XML_CHAR_ENCODING_EUC_JP:
1001 len = 45;
1002 break;
1003 }
1004 handler = xmlGetCharEncodingHandler(enc);
1005 if (handler == NULL) {
1006 /*
1007 * Default handlers.
1008 */
1009 switch (enc) {
1010 case XML_CHAR_ENCODING_ASCII:
1011 /* default encoding, no conversion should be needed */
1012 ctxt->charset = XML_CHAR_ENCODING_UTF8;
1013 return(0);
1014 case XML_CHAR_ENCODING_UTF16LE:
1015 break;
1016 case XML_CHAR_ENCODING_UTF16BE:
1017 break;
1018 case XML_CHAR_ENCODING_UCS4LE:
1019 __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
1020 "encoding not supported %s\n",
1021 BAD_CAST "USC4 little endian", NULL);
1022 break;
1023 case XML_CHAR_ENCODING_UCS4BE:
1024 __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
1025 "encoding not supported %s\n",
1026 BAD_CAST "USC4 big endian", NULL);
1027 break;
1028 case XML_CHAR_ENCODING_EBCDIC:
1029 __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
1030 "encoding not supported %s\n",
1031 BAD_CAST "EBCDIC", NULL);
1032 break;
1033 case XML_CHAR_ENCODING_UCS4_2143:
1034 __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
1035 "encoding not supported %s\n",
1036 BAD_CAST "UCS4 2143", NULL);
1037 break;
1038 case XML_CHAR_ENCODING_UCS4_3412:
1039 __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
1040 "encoding not supported %s\n",
1041 BAD_CAST "UCS4 3412", NULL);
1042 break;
1043 case XML_CHAR_ENCODING_UCS2:
1044 __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
1045 "encoding not supported %s\n",
1046 BAD_CAST "UCS2", NULL);
1047 break;
1048 case XML_CHAR_ENCODING_8859_1:
1049 case XML_CHAR_ENCODING_8859_2:
1050 case XML_CHAR_ENCODING_8859_3:
1051 case XML_CHAR_ENCODING_8859_4:
1052 case XML_CHAR_ENCODING_8859_5:
1053 case XML_CHAR_ENCODING_8859_6:
1054 case XML_CHAR_ENCODING_8859_7:
1055 case XML_CHAR_ENCODING_8859_8:
1056 case XML_CHAR_ENCODING_8859_9:
1057 /*
1058 * We used to keep the internal content in the
1059 * document encoding however this turns being unmaintainable
1060 * So xmlGetCharEncodingHandler() will return non-null
1061 * values for this now.
1062 */
1063 if ((ctxt->inputNr == 1) &&
1064 (ctxt->encoding == NULL) &&
1065 (ctxt->input != NULL) &&
1066 (ctxt->input->encoding != NULL)) {
1067 ctxt->encoding = xmlStrdup(ctxt->input->encoding);
1068 }
1069 ctxt->charset = enc;
1070 return(0);
1071 case XML_CHAR_ENCODING_2022_JP:
1072 __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
1073 "encoding not supported %s\n",
1074 BAD_CAST "ISO-2022-JP", NULL);
1075 break;
1076 case XML_CHAR_ENCODING_SHIFT_JIS:
1077 __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
1078 "encoding not supported %s\n",
1079 BAD_CAST "Shift_JIS", NULL);
1080 break;
1081 case XML_CHAR_ENCODING_EUC_JP:
1082 __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
1083 "encoding not supported %s\n",
1084 BAD_CAST "EUC-JP", NULL);
1085 break;
1086 default:
1087 break;
1088 }
1089 }
1090 /*
1091 * TODO: We could recover from errors in external entities if we
1092 * didn't stop the parser. But most callers of this function don't
1093 * check the return value.
1094 */
1095 if (handler == NULL) {
1096 xmlStopParser(ctxt);
1097 return(-1);
1098 }
1099 ctxt->charset = XML_CHAR_ENCODING_UTF8;
1100 ret = xmlSwitchToEncodingInt(ctxt, handler, len);
1101 if ((ret < 0) || (ctxt->errNo == XML_I18N_CONV_FAILED)) {
1102 /*
1103 * on encoding conversion errors, stop the parser
1104 */
1105 xmlStopParser(ctxt);
1106 ctxt->errNo = XML_I18N_CONV_FAILED;
1107 }
1108 return(ret);
1109 }
1110
1111 /**
1112 * xmlSwitchInputEncoding:
1113 * @ctxt: the parser context
1114 * @input: the input stream
1115 * @handler: the encoding handler
1116 * @len: the number of bytes to convert for the first line or -1
1117 *
1118 * change the input functions when discovering the character encoding
1119 * of a given entity.
1120 *
1121 * Returns 0 in case of success, -1 otherwise
1122 */
1123 static int
xmlSwitchInputEncodingInt(xmlParserCtxtPtr ctxt,xmlParserInputPtr input,xmlCharEncodingHandlerPtr handler,int len)1124 xmlSwitchInputEncodingInt(xmlParserCtxtPtr ctxt, xmlParserInputPtr input,
1125 xmlCharEncodingHandlerPtr handler, int len)
1126 {
1127 int nbchars;
1128
1129 if (handler == NULL)
1130 return (-1);
1131 if (input == NULL)
1132 return (-1);
1133 if (input->buf != NULL) {
1134 if (input->buf->encoder != NULL) {
1135 /*
1136 * Check in case the auto encoding detection triggered
1137 * in already.
1138 */
1139 if (input->buf->encoder == handler)
1140 return (0);
1141
1142 /*
1143 * "UTF-16" can be used for both LE and BE
1144 if ((!xmlStrncmp(BAD_CAST input->buf->encoder->name,
1145 BAD_CAST "UTF-16", 6)) &&
1146 (!xmlStrncmp(BAD_CAST handler->name,
1147 BAD_CAST "UTF-16", 6))) {
1148 return(0);
1149 }
1150 */
1151
1152 /*
1153 * Note: this is a bit dangerous, but that's what it
1154 * takes to use nearly compatible signature for different
1155 * encodings.
1156 *
1157 * FIXME: Encoders might buffer partial byte sequences, so
1158 * this probably can't work. We should return an error and
1159 * make sure that callers never try to switch the encoding
1160 * twice.
1161 */
1162 xmlCharEncCloseFunc(input->buf->encoder);
1163 input->buf->encoder = handler;
1164 return (0);
1165 }
1166 input->buf->encoder = handler;
1167
1168 /*
1169 * Is there already some content down the pipe to convert ?
1170 */
1171 if (xmlBufIsEmpty(input->buf->buffer) == 0) {
1172 int processed;
1173 unsigned int use;
1174
1175 /*
1176 * Specific handling of the Byte Order Mark for
1177 * UTF-16
1178 */
1179 if ((handler->name != NULL) &&
1180 (!strcmp(handler->name, "UTF-16LE") ||
1181 !strcmp(handler->name, "UTF-16")) &&
1182 (input->cur[0] == 0xFF) && (input->cur[1] == 0xFE)) {
1183 input->cur += 2;
1184 }
1185 if ((handler->name != NULL) &&
1186 (!strcmp(handler->name, "UTF-16BE")) &&
1187 (input->cur[0] == 0xFE) && (input->cur[1] == 0xFF)) {
1188 input->cur += 2;
1189 }
1190 /*
1191 * Errata on XML-1.0 June 20 2001
1192 * Specific handling of the Byte Order Mark for
1193 * UTF-8
1194 */
1195 if ((handler->name != NULL) &&
1196 (!strcmp(handler->name, "UTF-8")) &&
1197 (input->cur[0] == 0xEF) &&
1198 (input->cur[1] == 0xBB) && (input->cur[2] == 0xBF)) {
1199 input->cur += 3;
1200 }
1201
1202 /*
1203 * Shrink the current input buffer.
1204 * Move it as the raw buffer and create a new input buffer
1205 */
1206 processed = input->cur - input->base;
1207 xmlBufShrink(input->buf->buffer, processed);
1208 input->buf->raw = input->buf->buffer;
1209 input->buf->buffer = xmlBufCreate();
1210 input->buf->rawconsumed = processed;
1211 use = xmlBufUse(input->buf->raw);
1212
1213 if (ctxt->html) {
1214 /*
1215 * convert as much as possible of the buffer
1216 */
1217 nbchars = xmlCharEncInput(input->buf, 1);
1218 } else {
1219 /*
1220 * convert just enough to get
1221 * '<?xml version="1.0" encoding="xxx"?>'
1222 * parsed with the autodetected encoding
1223 * into the parser reading buffer.
1224 */
1225 nbchars = xmlCharEncFirstLineInput(input->buf, len);
1226 }
1227 xmlBufResetInput(input->buf->buffer, input);
1228 if (nbchars < 0) {
1229 xmlErrInternal(ctxt,
1230 "switching encoding: encoder error\n",
1231 NULL);
1232 return (-1);
1233 }
1234 input->buf->rawconsumed += use - xmlBufUse(input->buf->raw);
1235 }
1236 return (0);
1237 } else if (input->length == 0) {
1238 /*
1239 * When parsing a static memory array one must know the
1240 * size to be able to convert the buffer.
1241 */
1242 xmlErrInternal(ctxt, "switching encoding : no input\n", NULL);
1243 /*
1244 * Callers assume that the input buffer takes ownership of the
1245 * encoding handler. xmlCharEncCloseFunc frees unregistered
1246 * handlers and avoids a memory leak.
1247 */
1248 xmlCharEncCloseFunc(handler);
1249 return (-1);
1250 }
1251 /*
1252 * We should actually raise an error here, see issue #34.
1253 */
1254 xmlCharEncCloseFunc(handler);
1255 return (0);
1256 }
1257
1258 /**
1259 * xmlSwitchInputEncoding:
1260 * @ctxt: the parser context
1261 * @input: the input stream
1262 * @handler: the encoding handler
1263 *
1264 * change the input functions when discovering the character encoding
1265 * of a given entity.
1266 *
1267 * Returns 0 in case of success, -1 otherwise
1268 */
1269 int
xmlSwitchInputEncoding(xmlParserCtxtPtr ctxt,xmlParserInputPtr input,xmlCharEncodingHandlerPtr handler)1270 xmlSwitchInputEncoding(xmlParserCtxtPtr ctxt, xmlParserInputPtr input,
1271 xmlCharEncodingHandlerPtr handler) {
1272 return(xmlSwitchInputEncodingInt(ctxt, input, handler, -1));
1273 }
1274
1275 /**
1276 * xmlSwitchToEncodingInt:
1277 * @ctxt: the parser context
1278 * @handler: the encoding handler
1279 * @len: the length to convert or -1
1280 *
1281 * change the input functions when discovering the character encoding
1282 * of a given entity, and convert only @len bytes of the output, this
1283 * is needed on auto detect to allows any declared encoding later to
1284 * convert the actual content after the xmlDecl
1285 *
1286 * Returns 0 in case of success, -1 otherwise
1287 */
1288 static int
xmlSwitchToEncodingInt(xmlParserCtxtPtr ctxt,xmlCharEncodingHandlerPtr handler,int len)1289 xmlSwitchToEncodingInt(xmlParserCtxtPtr ctxt,
1290 xmlCharEncodingHandlerPtr handler, int len) {
1291 int ret = 0;
1292
1293 if (handler != NULL) {
1294 if (ctxt->input != NULL) {
1295 ret = xmlSwitchInputEncodingInt(ctxt, ctxt->input, handler, len);
1296 } else {
1297 xmlErrInternal(ctxt, "xmlSwitchToEncoding : no input\n",
1298 NULL);
1299 return(-1);
1300 }
1301 /*
1302 * The parsing is now done in UTF8 natively
1303 */
1304 ctxt->charset = XML_CHAR_ENCODING_UTF8;
1305 } else
1306 return(-1);
1307 return(ret);
1308 }
1309
1310 /**
1311 * xmlSwitchToEncoding:
1312 * @ctxt: the parser context
1313 * @handler: the encoding handler
1314 *
1315 * change the input functions when discovering the character encoding
1316 * of a given entity.
1317 *
1318 * Returns 0 in case of success, -1 otherwise
1319 */
1320 int
xmlSwitchToEncoding(xmlParserCtxtPtr ctxt,xmlCharEncodingHandlerPtr handler)1321 xmlSwitchToEncoding(xmlParserCtxtPtr ctxt, xmlCharEncodingHandlerPtr handler)
1322 {
1323 return (xmlSwitchToEncodingInt(ctxt, handler, -1));
1324 }
1325
1326 /************************************************************************
1327 * *
1328 * Commodity functions to handle entities processing *
1329 * *
1330 ************************************************************************/
1331
1332 /**
1333 * xmlFreeInputStream:
1334 * @input: an xmlParserInputPtr
1335 *
1336 * Free up an input stream.
1337 */
1338 void
xmlFreeInputStream(xmlParserInputPtr input)1339 xmlFreeInputStream(xmlParserInputPtr input) {
1340 if (input == NULL) return;
1341
1342 if (input->filename != NULL) xmlFree((char *) input->filename);
1343 if (input->directory != NULL) xmlFree((char *) input->directory);
1344 if (input->encoding != NULL) xmlFree((char *) input->encoding);
1345 if (input->version != NULL) xmlFree((char *) input->version);
1346 if ((input->free != NULL) && (input->base != NULL))
1347 input->free((xmlChar *) input->base);
1348 if (input->buf != NULL)
1349 xmlFreeParserInputBuffer(input->buf);
1350 xmlFree(input);
1351 }
1352
1353 /**
1354 * xmlNewInputStream:
1355 * @ctxt: an XML parser context
1356 *
1357 * Create a new input stream structure.
1358 *
1359 * Returns the new input stream or NULL
1360 */
1361 xmlParserInputPtr
xmlNewInputStream(xmlParserCtxtPtr ctxt)1362 xmlNewInputStream(xmlParserCtxtPtr ctxt) {
1363 xmlParserInputPtr input;
1364
1365 input = (xmlParserInputPtr) xmlMalloc(sizeof(xmlParserInput));
1366 if (input == NULL) {
1367 xmlErrMemory(ctxt, "couldn't allocate a new input stream\n");
1368 return(NULL);
1369 }
1370 memset(input, 0, sizeof(xmlParserInput));
1371 input->line = 1;
1372 input->col = 1;
1373 input->standalone = -1;
1374
1375 /*
1376 * If the context is NULL the id cannot be initialized, but that
1377 * should not happen while parsing which is the situation where
1378 * the id is actually needed.
1379 */
1380 if (ctxt != NULL)
1381 input->id = ctxt->input_id++;
1382
1383 return(input);
1384 }
1385
1386 /**
1387 * xmlNewIOInputStream:
1388 * @ctxt: an XML parser context
1389 * @input: an I/O Input
1390 * @enc: the charset encoding if known
1391 *
1392 * Create a new input stream structure encapsulating the @input into
1393 * a stream suitable for the parser.
1394 *
1395 * Returns the new input stream or NULL
1396 */
1397 xmlParserInputPtr
xmlNewIOInputStream(xmlParserCtxtPtr ctxt,xmlParserInputBufferPtr input,xmlCharEncoding enc)1398 xmlNewIOInputStream(xmlParserCtxtPtr ctxt, xmlParserInputBufferPtr input,
1399 xmlCharEncoding enc) {
1400 xmlParserInputPtr inputStream;
1401
1402 if (input == NULL) return(NULL);
1403 if (xmlParserDebugEntities)
1404 xmlGenericError(xmlGenericErrorContext, "new input from I/O\n");
1405 inputStream = xmlNewInputStream(ctxt);
1406 if (inputStream == NULL) {
1407 return(NULL);
1408 }
1409 inputStream->filename = NULL;
1410 inputStream->buf = input;
1411 xmlBufResetInput(inputStream->buf->buffer, inputStream);
1412
1413 if (enc != XML_CHAR_ENCODING_NONE) {
1414 xmlSwitchEncoding(ctxt, enc);
1415 }
1416
1417 return(inputStream);
1418 }
1419
1420 /**
1421 * xmlNewEntityInputStream:
1422 * @ctxt: an XML parser context
1423 * @entity: an Entity pointer
1424 *
1425 * Create a new input stream based on an xmlEntityPtr
1426 *
1427 * Returns the new input stream or NULL
1428 */
1429 xmlParserInputPtr
xmlNewEntityInputStream(xmlParserCtxtPtr ctxt,xmlEntityPtr entity)1430 xmlNewEntityInputStream(xmlParserCtxtPtr ctxt, xmlEntityPtr entity) {
1431 xmlParserInputPtr input;
1432
1433 if (entity == NULL) {
1434 xmlErrInternal(ctxt, "xmlNewEntityInputStream entity = NULL\n",
1435 NULL);
1436 return(NULL);
1437 }
1438 if (xmlParserDebugEntities)
1439 xmlGenericError(xmlGenericErrorContext,
1440 "new input from entity: %s\n", entity->name);
1441 if (entity->content == NULL) {
1442 switch (entity->etype) {
1443 case XML_EXTERNAL_GENERAL_UNPARSED_ENTITY:
1444 xmlErrInternal(ctxt, "Cannot parse entity %s\n",
1445 entity->name);
1446 break;
1447 case XML_EXTERNAL_GENERAL_PARSED_ENTITY:
1448 case XML_EXTERNAL_PARAMETER_ENTITY:
1449 return(xmlLoadExternalEntity((char *) entity->URI,
1450 (char *) entity->ExternalID, ctxt));
1451 case XML_INTERNAL_GENERAL_ENTITY:
1452 xmlErrInternal(ctxt,
1453 "Internal entity %s without content !\n",
1454 entity->name);
1455 break;
1456 case XML_INTERNAL_PARAMETER_ENTITY:
1457 xmlErrInternal(ctxt,
1458 "Internal parameter entity %s without content !\n",
1459 entity->name);
1460 break;
1461 case XML_INTERNAL_PREDEFINED_ENTITY:
1462 xmlErrInternal(ctxt,
1463 "Predefined entity %s without content !\n",
1464 entity->name);
1465 break;
1466 }
1467 return(NULL);
1468 }
1469 input = xmlNewInputStream(ctxt);
1470 if (input == NULL) {
1471 return(NULL);
1472 }
1473 if (entity->URI != NULL)
1474 input->filename = (char *) xmlStrdup((xmlChar *) entity->URI);
1475 input->base = entity->content;
1476 if (entity->length == 0)
1477 entity->length = xmlStrlen(entity->content);
1478 input->cur = entity->content;
1479 input->length = entity->length;
1480 input->end = &entity->content[input->length];
1481 return(input);
1482 }
1483
1484 /**
1485 * xmlNewStringInputStream:
1486 * @ctxt: an XML parser context
1487 * @buffer: an memory buffer
1488 *
1489 * Create a new input stream based on a memory buffer.
1490 * Returns the new input stream
1491 */
1492 xmlParserInputPtr
xmlNewStringInputStream(xmlParserCtxtPtr ctxt,const xmlChar * buffer)1493 xmlNewStringInputStream(xmlParserCtxtPtr ctxt, const xmlChar *buffer) {
1494 xmlParserInputPtr input;
1495
1496 if (buffer == NULL) {
1497 xmlErrInternal(ctxt, "xmlNewStringInputStream string = NULL\n",
1498 NULL);
1499 return(NULL);
1500 }
1501 if (xmlParserDebugEntities)
1502 xmlGenericError(xmlGenericErrorContext,
1503 "new fixed input: %.30s\n", buffer);
1504 input = xmlNewInputStream(ctxt);
1505 if (input == NULL) {
1506 xmlErrMemory(ctxt, "couldn't allocate a new input stream\n");
1507 return(NULL);
1508 }
1509 input->base = buffer;
1510 input->cur = buffer;
1511 input->length = xmlStrlen(buffer);
1512 input->end = &buffer[input->length];
1513 return(input);
1514 }
1515
1516 /**
1517 * xmlNewInputFromFile:
1518 * @ctxt: an XML parser context
1519 * @filename: the filename to use as entity
1520 *
1521 * Create a new input stream based on a file or an URL.
1522 *
1523 * Returns the new input stream or NULL in case of error
1524 */
1525 xmlParserInputPtr
xmlNewInputFromFile(xmlParserCtxtPtr ctxt,const char * filename)1526 xmlNewInputFromFile(xmlParserCtxtPtr ctxt, const char *filename) {
1527 xmlParserInputBufferPtr buf;
1528 xmlParserInputPtr inputStream;
1529 char *directory = NULL;
1530 xmlChar *URI = NULL;
1531
1532 if (xmlParserDebugEntities)
1533 xmlGenericError(xmlGenericErrorContext,
1534 "new input from file: %s\n", filename);
1535 if (ctxt == NULL) return(NULL);
1536 buf = xmlParserInputBufferCreateFilename(filename, XML_CHAR_ENCODING_NONE);
1537 if (buf == NULL) {
1538 if (filename == NULL)
1539 __xmlLoaderErr(ctxt,
1540 "failed to load external entity: NULL filename \n",
1541 NULL);
1542 else
1543 __xmlLoaderErr(ctxt, "failed to load external entity \"%s\"\n",
1544 (const char *) filename);
1545 return(NULL);
1546 }
1547
1548 inputStream = xmlNewInputStream(ctxt);
1549 if (inputStream == NULL)
1550 return(NULL);
1551
1552 inputStream->buf = buf;
1553 inputStream = xmlCheckHTTPInput(ctxt, inputStream);
1554 if (inputStream == NULL)
1555 return(NULL);
1556
1557 if (inputStream->filename == NULL)
1558 URI = xmlStrdup((xmlChar *) filename);
1559 else
1560 URI = xmlStrdup((xmlChar *) inputStream->filename);
1561 directory = xmlParserGetDirectory((const char *) URI);
1562 if (inputStream->filename != NULL) xmlFree((char *)inputStream->filename);
1563 inputStream->filename = (char *) xmlCanonicPath((const xmlChar *) URI);
1564 if (URI != NULL) xmlFree((char *) URI);
1565 inputStream->directory = directory;
1566
1567 xmlBufResetInput(inputStream->buf->buffer, inputStream);
1568 if ((ctxt->directory == NULL) && (directory != NULL))
1569 ctxt->directory = (char *) xmlStrdup((const xmlChar *) directory);
1570 return(inputStream);
1571 }
1572
1573 /************************************************************************
1574 * *
1575 * Commodity functions to handle parser contexts *
1576 * *
1577 ************************************************************************/
1578
1579 /**
1580 * xmlInitParserCtxt:
1581 * @ctxt: an XML parser context
1582 *
1583 * Initialize a parser context
1584 *
1585 * Returns 0 in case of success and -1 in case of error
1586 */
1587
1588 int
xmlInitParserCtxt(xmlParserCtxtPtr ctxt)1589 xmlInitParserCtxt(xmlParserCtxtPtr ctxt)
1590 {
1591 xmlParserInputPtr input;
1592
1593 if(ctxt==NULL) {
1594 xmlErrInternal(NULL, "Got NULL parser context\n", NULL);
1595 return(-1);
1596 }
1597
1598 xmlDefaultSAXHandlerInit();
1599
1600 if (ctxt->dict == NULL)
1601 ctxt->dict = xmlDictCreate();
1602 if (ctxt->dict == NULL) {
1603 xmlErrMemory(NULL, "cannot initialize parser context\n");
1604 return(-1);
1605 }
1606 xmlDictSetLimit(ctxt->dict, XML_MAX_DICTIONARY_LIMIT);
1607
1608 if (ctxt->sax == NULL)
1609 ctxt->sax = (xmlSAXHandler *) xmlMalloc(sizeof(xmlSAXHandler));
1610 if (ctxt->sax == NULL) {
1611 xmlErrMemory(NULL, "cannot initialize parser context\n");
1612 return(-1);
1613 }
1614 else
1615 xmlSAXVersion(ctxt->sax, 2);
1616
1617 ctxt->maxatts = 0;
1618 ctxt->atts = NULL;
1619 /* Allocate the Input stack */
1620 if (ctxt->inputTab == NULL) {
1621 ctxt->inputTab = (xmlParserInputPtr *)
1622 xmlMalloc(5 * sizeof(xmlParserInputPtr));
1623 ctxt->inputMax = 5;
1624 }
1625 if (ctxt->inputTab == NULL) {
1626 xmlErrMemory(NULL, "cannot initialize parser context\n");
1627 ctxt->inputNr = 0;
1628 ctxt->inputMax = 0;
1629 ctxt->input = NULL;
1630 return(-1);
1631 }
1632 while ((input = inputPop(ctxt)) != NULL) { /* Non consuming */
1633 xmlFreeInputStream(input);
1634 }
1635 ctxt->inputNr = 0;
1636 ctxt->input = NULL;
1637
1638 ctxt->version = NULL;
1639 ctxt->encoding = NULL;
1640 ctxt->standalone = -1;
1641 ctxt->hasExternalSubset = 0;
1642 ctxt->hasPErefs = 0;
1643 ctxt->html = 0;
1644 ctxt->external = 0;
1645 ctxt->instate = XML_PARSER_START;
1646 ctxt->token = 0;
1647 ctxt->directory = NULL;
1648
1649 /* Allocate the Node stack */
1650 if (ctxt->nodeTab == NULL) {
1651 ctxt->nodeTab = (xmlNodePtr *) xmlMalloc(10 * sizeof(xmlNodePtr));
1652 ctxt->nodeMax = 10;
1653 }
1654 if (ctxt->nodeTab == NULL) {
1655 xmlErrMemory(NULL, "cannot initialize parser context\n");
1656 ctxt->nodeNr = 0;
1657 ctxt->nodeMax = 0;
1658 ctxt->node = NULL;
1659 ctxt->inputNr = 0;
1660 ctxt->inputMax = 0;
1661 ctxt->input = NULL;
1662 return(-1);
1663 }
1664 ctxt->nodeNr = 0;
1665 ctxt->node = NULL;
1666
1667 /* Allocate the Name stack */
1668 if (ctxt->nameTab == NULL) {
1669 ctxt->nameTab = (const xmlChar **) xmlMalloc(10 * sizeof(xmlChar *));
1670 ctxt->nameMax = 10;
1671 }
1672 if (ctxt->nameTab == NULL) {
1673 xmlErrMemory(NULL, "cannot initialize parser context\n");
1674 ctxt->nodeNr = 0;
1675 ctxt->nodeMax = 0;
1676 ctxt->node = NULL;
1677 ctxt->inputNr = 0;
1678 ctxt->inputMax = 0;
1679 ctxt->input = NULL;
1680 ctxt->nameNr = 0;
1681 ctxt->nameMax = 0;
1682 ctxt->name = NULL;
1683 return(-1);
1684 }
1685 ctxt->nameNr = 0;
1686 ctxt->name = NULL;
1687
1688 /* Allocate the space stack */
1689 if (ctxt->spaceTab == NULL) {
1690 ctxt->spaceTab = (int *) xmlMalloc(10 * sizeof(int));
1691 ctxt->spaceMax = 10;
1692 }
1693 if (ctxt->spaceTab == NULL) {
1694 xmlErrMemory(NULL, "cannot initialize parser context\n");
1695 ctxt->nodeNr = 0;
1696 ctxt->nodeMax = 0;
1697 ctxt->node = NULL;
1698 ctxt->inputNr = 0;
1699 ctxt->inputMax = 0;
1700 ctxt->input = NULL;
1701 ctxt->nameNr = 0;
1702 ctxt->nameMax = 0;
1703 ctxt->name = NULL;
1704 ctxt->spaceNr = 0;
1705 ctxt->spaceMax = 0;
1706 ctxt->space = NULL;
1707 return(-1);
1708 }
1709 ctxt->spaceNr = 1;
1710 ctxt->spaceMax = 10;
1711 ctxt->spaceTab[0] = -1;
1712 ctxt->space = &ctxt->spaceTab[0];
1713 ctxt->userData = ctxt;
1714 ctxt->myDoc = NULL;
1715 ctxt->wellFormed = 1;
1716 ctxt->nsWellFormed = 1;
1717 ctxt->valid = 1;
1718 ctxt->loadsubset = xmlLoadExtDtdDefaultValue;
1719 if (ctxt->loadsubset) {
1720 ctxt->options |= XML_PARSE_DTDLOAD;
1721 }
1722 ctxt->validate = xmlDoValidityCheckingDefaultValue;
1723 ctxt->pedantic = xmlPedanticParserDefaultValue;
1724 if (ctxt->pedantic) {
1725 ctxt->options |= XML_PARSE_PEDANTIC;
1726 }
1727 ctxt->linenumbers = xmlLineNumbersDefaultValue;
1728 ctxt->keepBlanks = xmlKeepBlanksDefaultValue;
1729 if (ctxt->keepBlanks == 0) {
1730 ctxt->sax->ignorableWhitespace = xmlSAX2IgnorableWhitespace;
1731 ctxt->options |= XML_PARSE_NOBLANKS;
1732 }
1733
1734 ctxt->vctxt.finishDtd = XML_CTXT_FINISH_DTD_0;
1735 ctxt->vctxt.userData = ctxt;
1736 ctxt->vctxt.error = xmlParserValidityError;
1737 ctxt->vctxt.warning = xmlParserValidityWarning;
1738 if (ctxt->validate) {
1739 if (xmlGetWarningsDefaultValue == 0)
1740 ctxt->vctxt.warning = NULL;
1741 else
1742 ctxt->vctxt.warning = xmlParserValidityWarning;
1743 ctxt->vctxt.nodeMax = 0;
1744 ctxt->options |= XML_PARSE_DTDVALID;
1745 }
1746 ctxt->replaceEntities = xmlSubstituteEntitiesDefaultValue;
1747 if (ctxt->replaceEntities) {
1748 ctxt->options |= XML_PARSE_NOENT;
1749 }
1750 ctxt->record_info = 0;
1751 ctxt->checkIndex = 0;
1752 ctxt->inSubset = 0;
1753 ctxt->errNo = XML_ERR_OK;
1754 ctxt->depth = 0;
1755 ctxt->charset = XML_CHAR_ENCODING_UTF8;
1756 ctxt->catalogs = NULL;
1757 ctxt->nbentities = 0;
1758 ctxt->sizeentities = 0;
1759 ctxt->sizeentcopy = 0;
1760 ctxt->input_id = 1;
1761 xmlInitNodeInfoSeq(&ctxt->node_seq);
1762 return(0);
1763 }
1764
1765 /**
1766 * xmlFreeParserCtxt:
1767 * @ctxt: an XML parser context
1768 *
1769 * Free all the memory used by a parser context. However the parsed
1770 * document in ctxt->myDoc is not freed.
1771 */
1772
1773 void
xmlFreeParserCtxt(xmlParserCtxtPtr ctxt)1774 xmlFreeParserCtxt(xmlParserCtxtPtr ctxt)
1775 {
1776 xmlParserInputPtr input;
1777
1778 if (ctxt == NULL) return;
1779
1780 while ((input = inputPop(ctxt)) != NULL) { /* Non consuming */
1781 xmlFreeInputStream(input);
1782 }
1783 if (ctxt->spaceTab != NULL) xmlFree(ctxt->spaceTab);
1784 if (ctxt->nameTab != NULL) xmlFree((xmlChar * *)ctxt->nameTab);
1785 if (ctxt->nodeTab != NULL) xmlFree(ctxt->nodeTab);
1786 if (ctxt->nodeInfoTab != NULL) xmlFree(ctxt->nodeInfoTab);
1787 if (ctxt->inputTab != NULL) xmlFree(ctxt->inputTab);
1788 if (ctxt->version != NULL) xmlFree((char *) ctxt->version);
1789 if (ctxt->encoding != NULL) xmlFree((char *) ctxt->encoding);
1790 if (ctxt->extSubURI != NULL) xmlFree((char *) ctxt->extSubURI);
1791 if (ctxt->extSubSystem != NULL) xmlFree((char *) ctxt->extSubSystem);
1792 #ifdef LIBXML_SAX1_ENABLED
1793 if ((ctxt->sax != NULL) &&
1794 (ctxt->sax != (xmlSAXHandlerPtr) &xmlDefaultSAXHandler))
1795 #else
1796 if (ctxt->sax != NULL)
1797 #endif /* LIBXML_SAX1_ENABLED */
1798 xmlFree(ctxt->sax);
1799 if (ctxt->directory != NULL) xmlFree((char *) ctxt->directory);
1800 if (ctxt->vctxt.nodeTab != NULL) xmlFree(ctxt->vctxt.nodeTab);
1801 if (ctxt->atts != NULL) xmlFree((xmlChar * *)ctxt->atts);
1802 if (ctxt->dict != NULL) xmlDictFree(ctxt->dict);
1803 if (ctxt->nsTab != NULL) xmlFree((char *) ctxt->nsTab);
1804 if (ctxt->pushTab != NULL) xmlFree(ctxt->pushTab);
1805 if (ctxt->attallocs != NULL) xmlFree(ctxt->attallocs);
1806 if (ctxt->attsDefault != NULL)
1807 xmlHashFree(ctxt->attsDefault, xmlHashDefaultDeallocator);
1808 if (ctxt->attsSpecial != NULL)
1809 xmlHashFree(ctxt->attsSpecial, NULL);
1810 if (ctxt->freeElems != NULL) {
1811 xmlNodePtr cur, next;
1812
1813 cur = ctxt->freeElems;
1814 while (cur != NULL) {
1815 next = cur->next;
1816 xmlFree(cur);
1817 cur = next;
1818 }
1819 }
1820 if (ctxt->freeAttrs != NULL) {
1821 xmlAttrPtr cur, next;
1822
1823 cur = ctxt->freeAttrs;
1824 while (cur != NULL) {
1825 next = cur->next;
1826 xmlFree(cur);
1827 cur = next;
1828 }
1829 }
1830 /*
1831 * cleanup the error strings
1832 */
1833 if (ctxt->lastError.message != NULL)
1834 xmlFree(ctxt->lastError.message);
1835 if (ctxt->lastError.file != NULL)
1836 xmlFree(ctxt->lastError.file);
1837 if (ctxt->lastError.str1 != NULL)
1838 xmlFree(ctxt->lastError.str1);
1839 if (ctxt->lastError.str2 != NULL)
1840 xmlFree(ctxt->lastError.str2);
1841 if (ctxt->lastError.str3 != NULL)
1842 xmlFree(ctxt->lastError.str3);
1843
1844 #ifdef LIBXML_CATALOG_ENABLED
1845 if (ctxt->catalogs != NULL)
1846 xmlCatalogFreeLocal(ctxt->catalogs);
1847 #endif
1848 xmlFree(ctxt);
1849 }
1850
1851 /**
1852 * xmlNewParserCtxt:
1853 *
1854 * Allocate and initialize a new parser context.
1855 *
1856 * Returns the xmlParserCtxtPtr or NULL
1857 */
1858
1859 xmlParserCtxtPtr
xmlNewParserCtxt(void)1860 xmlNewParserCtxt(void)
1861 {
1862 xmlParserCtxtPtr ctxt;
1863
1864 ctxt = (xmlParserCtxtPtr) xmlMalloc(sizeof(xmlParserCtxt));
1865 if (ctxt == NULL) {
1866 xmlErrMemory(NULL, "cannot allocate parser context\n");
1867 return(NULL);
1868 }
1869 memset(ctxt, 0, sizeof(xmlParserCtxt));
1870 if (xmlInitParserCtxt(ctxt) < 0) {
1871 xmlFreeParserCtxt(ctxt);
1872 return(NULL);
1873 }
1874 return(ctxt);
1875 }
1876
1877 /************************************************************************
1878 * *
1879 * Handling of node information *
1880 * *
1881 ************************************************************************/
1882
1883 /**
1884 * xmlClearParserCtxt:
1885 * @ctxt: an XML parser context
1886 *
1887 * Clear (release owned resources) and reinitialize a parser context
1888 */
1889
1890 void
xmlClearParserCtxt(xmlParserCtxtPtr ctxt)1891 xmlClearParserCtxt(xmlParserCtxtPtr ctxt)
1892 {
1893 if (ctxt==NULL)
1894 return;
1895 xmlClearNodeInfoSeq(&ctxt->node_seq);
1896 xmlCtxtReset(ctxt);
1897 }
1898
1899
1900 /**
1901 * xmlParserFindNodeInfo:
1902 * @ctx: an XML parser context
1903 * @node: an XML node within the tree
1904 *
1905 * Find the parser node info struct for a given node
1906 *
1907 * Returns an xmlParserNodeInfo block pointer or NULL
1908 */
1909 const xmlParserNodeInfo *
xmlParserFindNodeInfo(const xmlParserCtxtPtr ctx,const xmlNodePtr node)1910 xmlParserFindNodeInfo(const xmlParserCtxtPtr ctx, const xmlNodePtr node)
1911 {
1912 unsigned long pos;
1913
1914 if ((ctx == NULL) || (node == NULL))
1915 return (NULL);
1916 /* Find position where node should be at */
1917 pos = xmlParserFindNodeInfoIndex(&ctx->node_seq, node);
1918 if (pos < ctx->node_seq.length
1919 && ctx->node_seq.buffer[pos].node == node)
1920 return &ctx->node_seq.buffer[pos];
1921 else
1922 return NULL;
1923 }
1924
1925
1926 /**
1927 * xmlInitNodeInfoSeq:
1928 * @seq: a node info sequence pointer
1929 *
1930 * -- Initialize (set to initial state) node info sequence
1931 */
1932 void
xmlInitNodeInfoSeq(xmlParserNodeInfoSeqPtr seq)1933 xmlInitNodeInfoSeq(xmlParserNodeInfoSeqPtr seq)
1934 {
1935 if (seq == NULL)
1936 return;
1937 seq->length = 0;
1938 seq->maximum = 0;
1939 seq->buffer = NULL;
1940 }
1941
1942 /**
1943 * xmlClearNodeInfoSeq:
1944 * @seq: a node info sequence pointer
1945 *
1946 * -- Clear (release memory and reinitialize) node
1947 * info sequence
1948 */
1949 void
xmlClearNodeInfoSeq(xmlParserNodeInfoSeqPtr seq)1950 xmlClearNodeInfoSeq(xmlParserNodeInfoSeqPtr seq)
1951 {
1952 if (seq == NULL)
1953 return;
1954 if (seq->buffer != NULL)
1955 xmlFree(seq->buffer);
1956 xmlInitNodeInfoSeq(seq);
1957 }
1958
1959 /**
1960 * xmlParserFindNodeInfoIndex:
1961 * @seq: a node info sequence pointer
1962 * @node: an XML node pointer
1963 *
1964 *
1965 * xmlParserFindNodeInfoIndex : Find the index that the info record for
1966 * the given node is or should be at in a sorted sequence
1967 *
1968 * Returns a long indicating the position of the record
1969 */
1970 unsigned long
xmlParserFindNodeInfoIndex(const xmlParserNodeInfoSeqPtr seq,const xmlNodePtr node)1971 xmlParserFindNodeInfoIndex(const xmlParserNodeInfoSeqPtr seq,
1972 const xmlNodePtr node)
1973 {
1974 unsigned long upper, lower, middle;
1975 int found = 0;
1976
1977 if ((seq == NULL) || (node == NULL))
1978 return ((unsigned long) -1);
1979
1980 /* Do a binary search for the key */
1981 lower = 1;
1982 upper = seq->length;
1983 middle = 0;
1984 while (lower <= upper && !found) {
1985 middle = lower + (upper - lower) / 2;
1986 if (node == seq->buffer[middle - 1].node)
1987 found = 1;
1988 else if (node < seq->buffer[middle - 1].node)
1989 upper = middle - 1;
1990 else
1991 lower = middle + 1;
1992 }
1993
1994 /* Return position */
1995 if (middle == 0 || seq->buffer[middle - 1].node < node)
1996 return middle;
1997 else
1998 return middle - 1;
1999 }
2000
2001
2002 /**
2003 * xmlParserAddNodeInfo:
2004 * @ctxt: an XML parser context
2005 * @info: a node info sequence pointer
2006 *
2007 * Insert node info record into the sorted sequence
2008 */
2009 void
xmlParserAddNodeInfo(xmlParserCtxtPtr ctxt,const xmlParserNodeInfoPtr info)2010 xmlParserAddNodeInfo(xmlParserCtxtPtr ctxt,
2011 const xmlParserNodeInfoPtr info)
2012 {
2013 unsigned long pos;
2014
2015 if ((ctxt == NULL) || (info == NULL)) return;
2016
2017 /* Find pos and check to see if node is already in the sequence */
2018 pos = xmlParserFindNodeInfoIndex(&ctxt->node_seq, (xmlNodePtr)
2019 info->node);
2020
2021 if ((pos < ctxt->node_seq.length) &&
2022 (ctxt->node_seq.buffer != NULL) &&
2023 (ctxt->node_seq.buffer[pos].node == info->node)) {
2024 ctxt->node_seq.buffer[pos] = *info;
2025 }
2026
2027 /* Otherwise, we need to add new node to buffer */
2028 else {
2029 if ((ctxt->node_seq.length + 1 > ctxt->node_seq.maximum) ||
2030 (ctxt->node_seq.buffer == NULL)) {
2031 xmlParserNodeInfo *tmp_buffer;
2032 unsigned int byte_size;
2033
2034 if (ctxt->node_seq.maximum == 0)
2035 ctxt->node_seq.maximum = 2;
2036 byte_size = (sizeof(*ctxt->node_seq.buffer) *
2037 (2 * ctxt->node_seq.maximum));
2038
2039 if (ctxt->node_seq.buffer == NULL)
2040 tmp_buffer = (xmlParserNodeInfo *) xmlMalloc(byte_size);
2041 else
2042 tmp_buffer =
2043 (xmlParserNodeInfo *) xmlRealloc(ctxt->node_seq.buffer,
2044 byte_size);
2045
2046 if (tmp_buffer == NULL) {
2047 xmlErrMemory(ctxt, "failed to allocate buffer\n");
2048 return;
2049 }
2050 ctxt->node_seq.buffer = tmp_buffer;
2051 ctxt->node_seq.maximum *= 2;
2052 }
2053
2054 /* If position is not at end, move elements out of the way */
2055 if (pos != ctxt->node_seq.length) {
2056 unsigned long i;
2057
2058 for (i = ctxt->node_seq.length; i > pos; i--)
2059 ctxt->node_seq.buffer[i] = ctxt->node_seq.buffer[i - 1];
2060 }
2061
2062 /* Copy element and increase length */
2063 ctxt->node_seq.buffer[pos] = *info;
2064 ctxt->node_seq.length++;
2065 }
2066 }
2067
2068 /************************************************************************
2069 * *
2070 * Defaults settings *
2071 * *
2072 ************************************************************************/
2073 /**
2074 * xmlPedanticParserDefault:
2075 * @val: int 0 or 1
2076 *
2077 * Set and return the previous value for enabling pedantic warnings.
2078 *
2079 * Returns the last value for 0 for no substitution, 1 for substitution.
2080 */
2081
2082 int
xmlPedanticParserDefault(int val)2083 xmlPedanticParserDefault(int val) {
2084 int old = xmlPedanticParserDefaultValue;
2085
2086 xmlPedanticParserDefaultValue = val;
2087 return(old);
2088 }
2089
2090 /**
2091 * xmlLineNumbersDefault:
2092 * @val: int 0 or 1
2093 *
2094 * Set and return the previous value for enabling line numbers in elements
2095 * contents. This may break on old application and is turned off by default.
2096 *
2097 * Returns the last value for 0 for no substitution, 1 for substitution.
2098 */
2099
2100 int
xmlLineNumbersDefault(int val)2101 xmlLineNumbersDefault(int val) {
2102 int old = xmlLineNumbersDefaultValue;
2103
2104 xmlLineNumbersDefaultValue = val;
2105 return(old);
2106 }
2107
2108 /**
2109 * xmlSubstituteEntitiesDefault:
2110 * @val: int 0 or 1
2111 *
2112 * Set and return the previous value for default entity support.
2113 * Initially the parser always keep entity references instead of substituting
2114 * entity values in the output. This function has to be used to change the
2115 * default parser behavior
2116 * SAX::substituteEntities() has to be used for changing that on a file by
2117 * file basis.
2118 *
2119 * Returns the last value for 0 for no substitution, 1 for substitution.
2120 */
2121
2122 int
xmlSubstituteEntitiesDefault(int val)2123 xmlSubstituteEntitiesDefault(int val) {
2124 int old = xmlSubstituteEntitiesDefaultValue;
2125
2126 xmlSubstituteEntitiesDefaultValue = val;
2127 return(old);
2128 }
2129
2130 /**
2131 * xmlKeepBlanksDefault:
2132 * @val: int 0 or 1
2133 *
2134 * Set and return the previous value for default blanks text nodes support.
2135 * The 1.x version of the parser used an heuristic to try to detect
2136 * ignorable white spaces. As a result the SAX callback was generating
2137 * xmlSAX2IgnorableWhitespace() callbacks instead of characters() one, and when
2138 * using the DOM output text nodes containing those blanks were not generated.
2139 * The 2.x and later version will switch to the XML standard way and
2140 * ignorableWhitespace() are only generated when running the parser in
2141 * validating mode and when the current element doesn't allow CDATA or
2142 * mixed content.
2143 * This function is provided as a way to force the standard behavior
2144 * on 1.X libs and to switch back to the old mode for compatibility when
2145 * running 1.X client code on 2.X . Upgrade of 1.X code should be done
2146 * by using xmlIsBlankNode() commodity function to detect the "empty"
2147 * nodes generated.
2148 * This value also affect autogeneration of indentation when saving code
2149 * if blanks sections are kept, indentation is not generated.
2150 *
2151 * Returns the last value for 0 for no substitution, 1 for substitution.
2152 */
2153
2154 int
xmlKeepBlanksDefault(int val)2155 xmlKeepBlanksDefault(int val) {
2156 int old = xmlKeepBlanksDefaultValue;
2157
2158 xmlKeepBlanksDefaultValue = val;
2159 if (!val) xmlIndentTreeOutput = 1;
2160 return(old);
2161 }
2162
2163 #define bottom_parserInternals
2164 #include "elfgcchack.h"
2165