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 <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 |
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 informations |
109 | * |
110 | * Handle a redefinition of attribute error |
111 | */ |
112 | void |
113 | xmlErrMemory(xmlParserCtxtPtr ctxt, const char *) |
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 |
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 informations |
169 | * |
170 | * Handle an internal error |
171 | */ |
172 | static void LIBXML_ATTR_FORMAT(2,0) |
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) |
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 |
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 |
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 |
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 |
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 |
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 |
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 | |
523 | ctxt->nbChars++; |
524 | } else { |
525 | /* |
526 | * Assume it's a fixed length encoding (1) with |
527 | * a compatible encoding for the ASCII set, since |
528 | * XML constructs only use < 128 chars |
529 | */ |
530 | |
531 | if (*(ctxt->input->cur) == '\n') { |
532 | ctxt->input->line++; ctxt->input->col = 1; |
533 | } else |
534 | ctxt->input->col++; |
535 | ctxt->input->cur++; |
536 | ctxt->nbChars++; |
537 | } |
538 | if (*ctxt->input->cur == 0) |
539 | xmlParserInputGrow(ctxt->input, INPUT_CHUNK); |
540 | return; |
541 | encoding_error: |
542 | /* |
543 | * If we detect an UTF8 error that probably mean that the |
544 | * input encoding didn't get properly advertised in the |
545 | * declaration header. Report the error and switch the encoding |
546 | * to ISO-Latin-1 (if you don't like this policy, just declare the |
547 | * encoding !) |
548 | */ |
549 | if ((ctxt == NULL) || (ctxt->input == NULL) || |
550 | (ctxt->input->end - ctxt->input->cur < 4)) { |
551 | __xmlErrEncoding(ctxt, XML_ERR_INVALID_CHAR, |
552 | "Input is not proper UTF-8, indicate encoding !\n" , |
553 | NULL, NULL); |
554 | } else { |
555 | char buffer[150]; |
556 | |
557 | snprintf(buffer, 149, "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n" , |
558 | ctxt->input->cur[0], ctxt->input->cur[1], |
559 | ctxt->input->cur[2], ctxt->input->cur[3]); |
560 | __xmlErrEncoding(ctxt, XML_ERR_INVALID_CHAR, |
561 | "Input is not proper UTF-8, indicate encoding !\n%s" , |
562 | BAD_CAST buffer, NULL); |
563 | } |
564 | ctxt->charset = XML_CHAR_ENCODING_8859_1; |
565 | ctxt->input->cur++; |
566 | return; |
567 | } |
568 | |
569 | /** |
570 | * xmlCurrentChar: |
571 | * @ctxt: the XML parser context |
572 | * @len: pointer to the length of the char read |
573 | * |
574 | * The current char value, if using UTF-8 this may actually span multiple |
575 | * bytes in the input buffer. Implement the end of line normalization: |
576 | * 2.11 End-of-Line Handling |
577 | * Wherever an external parsed entity or the literal entity value |
578 | * of an internal parsed entity contains either the literal two-character |
579 | * sequence "#xD#xA" or a standalone literal #xD, an XML processor |
580 | * must pass to the application the single character #xA. |
581 | * This behavior can conveniently be produced by normalizing all |
582 | * line breaks to #xA on input, before parsing.) |
583 | * |
584 | * Returns the current char value and its length |
585 | */ |
586 | |
587 | int |
588 | xmlCurrentChar(xmlParserCtxtPtr ctxt, int *len) { |
589 | if ((ctxt == NULL) || (len == NULL) || (ctxt->input == NULL)) return(0); |
590 | if (ctxt->instate == XML_PARSER_EOF) |
591 | return(0); |
592 | |
593 | if ((*ctxt->input->cur >= 0x20) && (*ctxt->input->cur <= 0x7F)) { |
594 | *len = 1; |
595 | return((int) *ctxt->input->cur); |
596 | } |
597 | if (ctxt->charset == XML_CHAR_ENCODING_UTF8) { |
598 | /* |
599 | * We are supposed to handle UTF8, check it's valid |
600 | * From rfc2044: encoding of the Unicode values on UTF-8: |
601 | * |
602 | * UCS-4 range (hex.) UTF-8 octet sequence (binary) |
603 | * 0000 0000-0000 007F 0xxxxxxx |
604 | * 0000 0080-0000 07FF 110xxxxx 10xxxxxx |
605 | * 0000 0800-0000 FFFF 1110xxxx 10xxxxxx 10xxxxxx |
606 | * |
607 | * Check for the 0x110000 limit too |
608 | */ |
609 | const unsigned char *cur = ctxt->input->cur; |
610 | unsigned char c; |
611 | unsigned int val; |
612 | |
613 | c = *cur; |
614 | if (c & 0x80) { |
615 | if (((c & 0x40) == 0) || (c == 0xC0)) |
616 | goto encoding_error; |
617 | if (cur[1] == 0) { |
618 | xmlParserInputGrow(ctxt->input, INPUT_CHUNK); |
619 | cur = ctxt->input->cur; |
620 | } |
621 | if ((cur[1] & 0xc0) != 0x80) |
622 | goto encoding_error; |
623 | if ((c & 0xe0) == 0xe0) { |
624 | if (cur[2] == 0) { |
625 | xmlParserInputGrow(ctxt->input, INPUT_CHUNK); |
626 | cur = ctxt->input->cur; |
627 | } |
628 | if ((cur[2] & 0xc0) != 0x80) |
629 | goto encoding_error; |
630 | if ((c & 0xf0) == 0xf0) { |
631 | if (cur[3] == 0) { |
632 | xmlParserInputGrow(ctxt->input, INPUT_CHUNK); |
633 | cur = ctxt->input->cur; |
634 | } |
635 | if (((c & 0xf8) != 0xf0) || |
636 | ((cur[3] & 0xc0) != 0x80)) |
637 | goto encoding_error; |
638 | /* 4-byte code */ |
639 | *len = 4; |
640 | val = (cur[0] & 0x7) << 18; |
641 | val |= (cur[1] & 0x3f) << 12; |
642 | val |= (cur[2] & 0x3f) << 6; |
643 | val |= cur[3] & 0x3f; |
644 | if (val < 0x10000) |
645 | goto encoding_error; |
646 | } else { |
647 | /* 3-byte code */ |
648 | *len = 3; |
649 | val = (cur[0] & 0xf) << 12; |
650 | val |= (cur[1] & 0x3f) << 6; |
651 | val |= cur[2] & 0x3f; |
652 | if (val < 0x800) |
653 | goto encoding_error; |
654 | } |
655 | } else { |
656 | /* 2-byte code */ |
657 | *len = 2; |
658 | val = (cur[0] & 0x1f) << 6; |
659 | val |= cur[1] & 0x3f; |
660 | if (val < 0x80) |
661 | goto encoding_error; |
662 | } |
663 | if (!IS_CHAR(val)) { |
664 | xmlErrEncodingInt(ctxt, XML_ERR_INVALID_CHAR, |
665 | "Char 0x%X out of allowed range\n" , val); |
666 | } |
667 | return(val); |
668 | } else { |
669 | /* 1-byte code */ |
670 | *len = 1; |
671 | if (*ctxt->input->cur == 0) |
672 | xmlParserInputGrow(ctxt->input, INPUT_CHUNK); |
673 | if ((*ctxt->input->cur == 0) && |
674 | (ctxt->input->end > ctxt->input->cur)) { |
675 | xmlErrEncodingInt(ctxt, XML_ERR_INVALID_CHAR, |
676 | "Char 0x0 out of allowed range\n" , 0); |
677 | } |
678 | if (*ctxt->input->cur == 0xD) { |
679 | if (ctxt->input->cur[1] == 0xA) { |
680 | ctxt->nbChars++; |
681 | ctxt->input->cur++; |
682 | } |
683 | return(0xA); |
684 | } |
685 | return((int) *ctxt->input->cur); |
686 | } |
687 | } |
688 | /* |
689 | * Assume it's a fixed length encoding (1) with |
690 | * a compatible encoding for the ASCII set, since |
691 | * XML constructs only use < 128 chars |
692 | */ |
693 | *len = 1; |
694 | if (*ctxt->input->cur == 0xD) { |
695 | if (ctxt->input->cur[1] == 0xA) { |
696 | ctxt->nbChars++; |
697 | ctxt->input->cur++; |
698 | } |
699 | return(0xA); |
700 | } |
701 | return((int) *ctxt->input->cur); |
702 | encoding_error: |
703 | /* |
704 | * An encoding problem may arise from a truncated input buffer |
705 | * splitting a character in the middle. In that case do not raise |
706 | * an error but return 0 to endicate an end of stream problem |
707 | */ |
708 | if (ctxt->input->end - ctxt->input->cur < 4) { |
709 | *len = 0; |
710 | return(0); |
711 | } |
712 | |
713 | /* |
714 | * If we detect an UTF8 error that probably mean that the |
715 | * input encoding didn't get properly advertised in the |
716 | * declaration header. Report the error and switch the encoding |
717 | * to ISO-Latin-1 (if you don't like this policy, just declare the |
718 | * encoding !) |
719 | */ |
720 | { |
721 | char buffer[150]; |
722 | |
723 | snprintf(&buffer[0], 149, "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n" , |
724 | ctxt->input->cur[0], ctxt->input->cur[1], |
725 | ctxt->input->cur[2], ctxt->input->cur[3]); |
726 | __xmlErrEncoding(ctxt, XML_ERR_INVALID_CHAR, |
727 | "Input is not proper UTF-8, indicate encoding !\n%s" , |
728 | BAD_CAST buffer, NULL); |
729 | } |
730 | ctxt->charset = XML_CHAR_ENCODING_8859_1; |
731 | *len = 1; |
732 | return((int) *ctxt->input->cur); |
733 | } |
734 | |
735 | /** |
736 | * xmlStringCurrentChar: |
737 | * @ctxt: the XML parser context |
738 | * @cur: pointer to the beginning of the char |
739 | * @len: pointer to the length of the char read |
740 | * |
741 | * The current char value, if using UTF-8 this may actually span multiple |
742 | * bytes in the input buffer. |
743 | * |
744 | * Returns the current char value and its length |
745 | */ |
746 | |
747 | int |
748 | xmlStringCurrentChar(xmlParserCtxtPtr ctxt, const xmlChar * cur, int *len) |
749 | { |
750 | if ((len == NULL) || (cur == NULL)) return(0); |
751 | if ((ctxt == NULL) || (ctxt->charset == XML_CHAR_ENCODING_UTF8)) { |
752 | /* |
753 | * We are supposed to handle UTF8, check it's valid |
754 | * From rfc2044: encoding of the Unicode values on UTF-8: |
755 | * |
756 | * UCS-4 range (hex.) UTF-8 octet sequence (binary) |
757 | * 0000 0000-0000 007F 0xxxxxxx |
758 | * 0000 0080-0000 07FF 110xxxxx 10xxxxxx |
759 | * 0000 0800-0000 FFFF 1110xxxx 10xxxxxx 10xxxxxx |
760 | * |
761 | * Check for the 0x110000 limit too |
762 | */ |
763 | unsigned char c; |
764 | unsigned int val; |
765 | |
766 | c = *cur; |
767 | if (c & 0x80) { |
768 | if ((cur[1] & 0xc0) != 0x80) |
769 | goto encoding_error; |
770 | if ((c & 0xe0) == 0xe0) { |
771 | |
772 | if ((cur[2] & 0xc0) != 0x80) |
773 | goto encoding_error; |
774 | if ((c & 0xf0) == 0xf0) { |
775 | if (((c & 0xf8) != 0xf0) || ((cur[3] & 0xc0) != 0x80)) |
776 | goto encoding_error; |
777 | /* 4-byte code */ |
778 | *len = 4; |
779 | val = (cur[0] & 0x7) << 18; |
780 | val |= (cur[1] & 0x3f) << 12; |
781 | val |= (cur[2] & 0x3f) << 6; |
782 | val |= cur[3] & 0x3f; |
783 | } else { |
784 | /* 3-byte code */ |
785 | *len = 3; |
786 | val = (cur[0] & 0xf) << 12; |
787 | val |= (cur[1] & 0x3f) << 6; |
788 | val |= cur[2] & 0x3f; |
789 | } |
790 | } else { |
791 | /* 2-byte code */ |
792 | *len = 2; |
793 | val = (cur[0] & 0x1f) << 6; |
794 | val |= cur[1] & 0x3f; |
795 | } |
796 | if (!IS_CHAR(val)) { |
797 | xmlErrEncodingInt(ctxt, XML_ERR_INVALID_CHAR, |
798 | "Char 0x%X out of allowed range\n" , val); |
799 | } |
800 | return (val); |
801 | } else { |
802 | /* 1-byte code */ |
803 | *len = 1; |
804 | return ((int) *cur); |
805 | } |
806 | } |
807 | /* |
808 | * Assume it's a fixed length encoding (1) with |
809 | * a compatible encoding for the ASCII set, since |
810 | * XML constructs only use < 128 chars |
811 | */ |
812 | *len = 1; |
813 | return ((int) *cur); |
814 | encoding_error: |
815 | |
816 | /* |
817 | * An encoding problem may arise from a truncated input buffer |
818 | * splitting a character in the middle. In that case do not raise |
819 | * an error but return 0 to endicate an end of stream problem |
820 | */ |
821 | if ((ctxt == NULL) || (ctxt->input == NULL) || |
822 | (ctxt->input->end - ctxt->input->cur < 4)) { |
823 | *len = 0; |
824 | return(0); |
825 | } |
826 | /* |
827 | * If we detect an UTF8 error that probably mean that the |
828 | * input encoding didn't get properly advertised in the |
829 | * declaration header. Report the error and switch the encoding |
830 | * to ISO-Latin-1 (if you don't like this policy, just declare the |
831 | * encoding !) |
832 | */ |
833 | { |
834 | char buffer[150]; |
835 | |
836 | snprintf(buffer, 149, "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n" , |
837 | ctxt->input->cur[0], ctxt->input->cur[1], |
838 | ctxt->input->cur[2], ctxt->input->cur[3]); |
839 | __xmlErrEncoding(ctxt, XML_ERR_INVALID_CHAR, |
840 | "Input is not proper UTF-8, indicate encoding !\n%s" , |
841 | BAD_CAST buffer, NULL); |
842 | } |
843 | *len = 1; |
844 | return ((int) *cur); |
845 | } |
846 | |
847 | /** |
848 | * xmlCopyCharMultiByte: |
849 | * @out: pointer to an array of xmlChar |
850 | * @val: the char value |
851 | * |
852 | * append the char value in the array |
853 | * |
854 | * Returns the number of xmlChar written |
855 | */ |
856 | int |
857 | xmlCopyCharMultiByte(xmlChar *out, int val) { |
858 | if (out == NULL) return(0); |
859 | /* |
860 | * We are supposed to handle UTF8, check it's valid |
861 | * From rfc2044: encoding of the Unicode values on UTF-8: |
862 | * |
863 | * UCS-4 range (hex.) UTF-8 octet sequence (binary) |
864 | * 0000 0000-0000 007F 0xxxxxxx |
865 | * 0000 0080-0000 07FF 110xxxxx 10xxxxxx |
866 | * 0000 0800-0000 FFFF 1110xxxx 10xxxxxx 10xxxxxx |
867 | */ |
868 | if (val >= 0x80) { |
869 | xmlChar *savedout = out; |
870 | int bits; |
871 | if (val < 0x800) { *out++= (val >> 6) | 0xC0; bits= 0; } |
872 | else if (val < 0x10000) { *out++= (val >> 12) | 0xE0; bits= 6;} |
873 | else if (val < 0x110000) { *out++= (val >> 18) | 0xF0; bits= 12; } |
874 | else { |
875 | xmlErrEncodingInt(NULL, XML_ERR_INVALID_CHAR, |
876 | "Internal error, xmlCopyCharMultiByte 0x%X out of bound\n" , |
877 | val); |
878 | return(0); |
879 | } |
880 | for ( ; bits >= 0; bits-= 6) |
881 | *out++= ((val >> bits) & 0x3F) | 0x80 ; |
882 | return (out - savedout); |
883 | } |
884 | *out = (xmlChar) val; |
885 | return 1; |
886 | } |
887 | |
888 | /** |
889 | * xmlCopyChar: |
890 | * @len: Ignored, compatibility |
891 | * @out: pointer to an array of xmlChar |
892 | * @val: the char value |
893 | * |
894 | * append the char value in the array |
895 | * |
896 | * Returns the number of xmlChar written |
897 | */ |
898 | |
899 | int |
900 | xmlCopyChar(int len ATTRIBUTE_UNUSED, xmlChar *out, int val) { |
901 | if (out == NULL) return(0); |
902 | /* the len parameter is ignored */ |
903 | if (val >= 0x80) { |
904 | return(xmlCopyCharMultiByte (out, val)); |
905 | } |
906 | *out = (xmlChar) val; |
907 | return 1; |
908 | } |
909 | |
910 | /************************************************************************ |
911 | * * |
912 | * Commodity functions to switch encodings * |
913 | * * |
914 | ************************************************************************/ |
915 | |
916 | static int |
917 | xmlSwitchToEncodingInt(xmlParserCtxtPtr ctxt, |
918 | xmlCharEncodingHandlerPtr handler, int len); |
919 | static int |
920 | xmlSwitchInputEncodingInt(xmlParserCtxtPtr ctxt, xmlParserInputPtr input, |
921 | xmlCharEncodingHandlerPtr handler, int len); |
922 | /** |
923 | * xmlSwitchEncoding: |
924 | * @ctxt: the parser context |
925 | * @enc: the encoding value (number) |
926 | * |
927 | * change the input functions when discovering the character encoding |
928 | * of a given entity. |
929 | * |
930 | * Returns 0 in case of success, -1 otherwise |
931 | */ |
932 | int |
933 | xmlSwitchEncoding(xmlParserCtxtPtr ctxt, xmlCharEncoding enc) |
934 | { |
935 | xmlCharEncodingHandlerPtr handler; |
936 | int len = -1; |
937 | int ret; |
938 | |
939 | if (ctxt == NULL) return(-1); |
940 | switch (enc) { |
941 | case XML_CHAR_ENCODING_ERROR: |
942 | __xmlErrEncoding(ctxt, XML_ERR_UNKNOWN_ENCODING, |
943 | "encoding unknown\n" , NULL, NULL); |
944 | return(-1); |
945 | case XML_CHAR_ENCODING_NONE: |
946 | /* let's assume it's UTF-8 without the XML decl */ |
947 | ctxt->charset = XML_CHAR_ENCODING_UTF8; |
948 | return(0); |
949 | case XML_CHAR_ENCODING_UTF8: |
950 | /* default encoding, no conversion should be needed */ |
951 | ctxt->charset = XML_CHAR_ENCODING_UTF8; |
952 | |
953 | /* |
954 | * Errata on XML-1.0 June 20 2001 |
955 | * Specific handling of the Byte Order Mark for |
956 | * UTF-8 |
957 | */ |
958 | if ((ctxt->input != NULL) && |
959 | (ctxt->input->cur[0] == 0xEF) && |
960 | (ctxt->input->cur[1] == 0xBB) && |
961 | (ctxt->input->cur[2] == 0xBF)) { |
962 | ctxt->input->cur += 3; |
963 | } |
964 | return(0); |
965 | case XML_CHAR_ENCODING_UTF16LE: |
966 | case XML_CHAR_ENCODING_UTF16BE: |
967 | /*The raw input characters are encoded |
968 | *in UTF-16. As we expect this function |
969 | *to be called after xmlCharEncInFunc, we expect |
970 | *ctxt->input->cur to contain UTF-8 encoded characters. |
971 | *So the raw UTF16 Byte Order Mark |
972 | *has also been converted into |
973 | *an UTF-8 BOM. Let's skip that BOM. |
974 | */ |
975 | if ((ctxt->input != NULL) && (ctxt->input->cur != NULL) && |
976 | (ctxt->input->cur[0] == 0xEF) && |
977 | (ctxt->input->cur[1] == 0xBB) && |
978 | (ctxt->input->cur[2] == 0xBF)) { |
979 | ctxt->input->cur += 3; |
980 | } |
981 | len = 90; |
982 | break; |
983 | case XML_CHAR_ENCODING_UCS2: |
984 | len = 90; |
985 | break; |
986 | case XML_CHAR_ENCODING_UCS4BE: |
987 | case XML_CHAR_ENCODING_UCS4LE: |
988 | case XML_CHAR_ENCODING_UCS4_2143: |
989 | case XML_CHAR_ENCODING_UCS4_3412: |
990 | len = 180; |
991 | break; |
992 | case XML_CHAR_ENCODING_EBCDIC: |
993 | case XML_CHAR_ENCODING_8859_1: |
994 | case XML_CHAR_ENCODING_8859_2: |
995 | case XML_CHAR_ENCODING_8859_3: |
996 | case XML_CHAR_ENCODING_8859_4: |
997 | case XML_CHAR_ENCODING_8859_5: |
998 | case XML_CHAR_ENCODING_8859_6: |
999 | case XML_CHAR_ENCODING_8859_7: |
1000 | case XML_CHAR_ENCODING_8859_8: |
1001 | case XML_CHAR_ENCODING_8859_9: |
1002 | case XML_CHAR_ENCODING_ASCII: |
1003 | case XML_CHAR_ENCODING_2022_JP: |
1004 | case XML_CHAR_ENCODING_SHIFT_JIS: |
1005 | case XML_CHAR_ENCODING_EUC_JP: |
1006 | len = 45; |
1007 | break; |
1008 | } |
1009 | handler = xmlGetCharEncodingHandler(enc); |
1010 | if (handler == NULL) { |
1011 | /* |
1012 | * Default handlers. |
1013 | */ |
1014 | switch (enc) { |
1015 | case XML_CHAR_ENCODING_ASCII: |
1016 | /* default encoding, no conversion should be needed */ |
1017 | ctxt->charset = XML_CHAR_ENCODING_UTF8; |
1018 | return(0); |
1019 | case XML_CHAR_ENCODING_UTF16LE: |
1020 | break; |
1021 | case XML_CHAR_ENCODING_UTF16BE: |
1022 | break; |
1023 | case XML_CHAR_ENCODING_UCS4LE: |
1024 | __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING, |
1025 | "encoding not supported %s\n" , |
1026 | BAD_CAST "USC4 little endian" , NULL); |
1027 | break; |
1028 | case XML_CHAR_ENCODING_UCS4BE: |
1029 | __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING, |
1030 | "encoding not supported %s\n" , |
1031 | BAD_CAST "USC4 big endian" , NULL); |
1032 | break; |
1033 | case XML_CHAR_ENCODING_EBCDIC: |
1034 | __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING, |
1035 | "encoding not supported %s\n" , |
1036 | BAD_CAST "EBCDIC" , NULL); |
1037 | break; |
1038 | case XML_CHAR_ENCODING_UCS4_2143: |
1039 | __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING, |
1040 | "encoding not supported %s\n" , |
1041 | BAD_CAST "UCS4 2143" , NULL); |
1042 | break; |
1043 | case XML_CHAR_ENCODING_UCS4_3412: |
1044 | __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING, |
1045 | "encoding not supported %s\n" , |
1046 | BAD_CAST "UCS4 3412" , NULL); |
1047 | break; |
1048 | case XML_CHAR_ENCODING_UCS2: |
1049 | __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING, |
1050 | "encoding not supported %s\n" , |
1051 | BAD_CAST "UCS2" , NULL); |
1052 | break; |
1053 | case XML_CHAR_ENCODING_8859_1: |
1054 | case XML_CHAR_ENCODING_8859_2: |
1055 | case XML_CHAR_ENCODING_8859_3: |
1056 | case XML_CHAR_ENCODING_8859_4: |
1057 | case XML_CHAR_ENCODING_8859_5: |
1058 | case XML_CHAR_ENCODING_8859_6: |
1059 | case XML_CHAR_ENCODING_8859_7: |
1060 | case XML_CHAR_ENCODING_8859_8: |
1061 | case XML_CHAR_ENCODING_8859_9: |
1062 | /* |
1063 | * We used to keep the internal content in the |
1064 | * document encoding however this turns being unmaintainable |
1065 | * So xmlGetCharEncodingHandler() will return non-null |
1066 | * values for this now. |
1067 | */ |
1068 | if ((ctxt->inputNr == 1) && |
1069 | (ctxt->encoding == NULL) && |
1070 | (ctxt->input != NULL) && |
1071 | (ctxt->input->encoding != NULL)) { |
1072 | ctxt->encoding = xmlStrdup(ctxt->input->encoding); |
1073 | } |
1074 | ctxt->charset = enc; |
1075 | return(0); |
1076 | case XML_CHAR_ENCODING_2022_JP: |
1077 | __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING, |
1078 | "encoding not supported %s\n" , |
1079 | BAD_CAST "ISO-2022-JP" , NULL); |
1080 | break; |
1081 | case XML_CHAR_ENCODING_SHIFT_JIS: |
1082 | __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING, |
1083 | "encoding not supported %s\n" , |
1084 | BAD_CAST "Shift_JIS" , NULL); |
1085 | break; |
1086 | case XML_CHAR_ENCODING_EUC_JP: |
1087 | __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING, |
1088 | "encoding not supported %s\n" , |
1089 | BAD_CAST "EUC-JP" , NULL); |
1090 | break; |
1091 | default: |
1092 | break; |
1093 | } |
1094 | } |
1095 | /* |
1096 | * TODO: We could recover from errors in external entites if we |
1097 | * didn't stop the parser. But most callers of this function don't |
1098 | * check the return value. |
1099 | */ |
1100 | if (handler == NULL) { |
1101 | xmlStopParser(ctxt); |
1102 | return(-1); |
1103 | } |
1104 | ctxt->charset = XML_CHAR_ENCODING_UTF8; |
1105 | ret = xmlSwitchToEncodingInt(ctxt, handler, len); |
1106 | if ((ret < 0) || (ctxt->errNo == XML_I18N_CONV_FAILED)) { |
1107 | /* |
1108 | * on encoding conversion errors, stop the parser |
1109 | */ |
1110 | xmlStopParser(ctxt); |
1111 | ctxt->errNo = XML_I18N_CONV_FAILED; |
1112 | } |
1113 | return(ret); |
1114 | } |
1115 | |
1116 | /** |
1117 | * xmlSwitchInputEncoding: |
1118 | * @ctxt: the parser context |
1119 | * @input: the input stream |
1120 | * @handler: the encoding handler |
1121 | * @len: the number of bytes to convert for the first line or -1 |
1122 | * |
1123 | * change the input functions when discovering the character encoding |
1124 | * of a given entity. |
1125 | * |
1126 | * Returns 0 in case of success, -1 otherwise |
1127 | */ |
1128 | static int |
1129 | xmlSwitchInputEncodingInt(xmlParserCtxtPtr ctxt, xmlParserInputPtr input, |
1130 | xmlCharEncodingHandlerPtr handler, int len) |
1131 | { |
1132 | int nbchars; |
1133 | |
1134 | if (handler == NULL) |
1135 | return (-1); |
1136 | if (input == NULL) |
1137 | return (-1); |
1138 | if (input->buf != NULL) { |
1139 | if (input->buf->encoder != NULL) { |
1140 | /* |
1141 | * Check in case the auto encoding detetection triggered |
1142 | * in already. |
1143 | */ |
1144 | if (input->buf->encoder == handler) |
1145 | return (0); |
1146 | |
1147 | /* |
1148 | * "UTF-16" can be used for both LE and BE |
1149 | if ((!xmlStrncmp(BAD_CAST input->buf->encoder->name, |
1150 | BAD_CAST "UTF-16", 6)) && |
1151 | (!xmlStrncmp(BAD_CAST handler->name, |
1152 | BAD_CAST "UTF-16", 6))) { |
1153 | return(0); |
1154 | } |
1155 | */ |
1156 | |
1157 | /* |
1158 | * Note: this is a bit dangerous, but that's what it |
1159 | * takes to use nearly compatible signature for different |
1160 | * encodings. |
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, 0); |
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 | return (-1); |
1244 | } |
1245 | return (0); |
1246 | } |
1247 | |
1248 | /** |
1249 | * xmlSwitchInputEncoding: |
1250 | * @ctxt: the parser context |
1251 | * @input: the input stream |
1252 | * @handler: the encoding handler |
1253 | * |
1254 | * change the input functions when discovering the character encoding |
1255 | * of a given entity. |
1256 | * |
1257 | * Returns 0 in case of success, -1 otherwise |
1258 | */ |
1259 | int |
1260 | xmlSwitchInputEncoding(xmlParserCtxtPtr ctxt, xmlParserInputPtr input, |
1261 | xmlCharEncodingHandlerPtr handler) { |
1262 | return(xmlSwitchInputEncodingInt(ctxt, input, handler, -1)); |
1263 | } |
1264 | |
1265 | /** |
1266 | * xmlSwitchToEncodingInt: |
1267 | * @ctxt: the parser context |
1268 | * @handler: the encoding handler |
1269 | * @len: the length to convert or -1 |
1270 | * |
1271 | * change the input functions when discovering the character encoding |
1272 | * of a given entity, and convert only @len bytes of the output, this |
1273 | * is needed on auto detect to allows any declared encoding later to |
1274 | * convert the actual content after the xmlDecl |
1275 | * |
1276 | * Returns 0 in case of success, -1 otherwise |
1277 | */ |
1278 | static int |
1279 | xmlSwitchToEncodingInt(xmlParserCtxtPtr ctxt, |
1280 | xmlCharEncodingHandlerPtr handler, int len) { |
1281 | int ret = 0; |
1282 | |
1283 | if (handler != NULL) { |
1284 | if (ctxt->input != NULL) { |
1285 | ret = xmlSwitchInputEncodingInt(ctxt, ctxt->input, handler, len); |
1286 | } else { |
1287 | xmlErrInternal(ctxt, "xmlSwitchToEncoding : no input\n" , |
1288 | NULL); |
1289 | return(-1); |
1290 | } |
1291 | /* |
1292 | * The parsing is now done in UTF8 natively |
1293 | */ |
1294 | ctxt->charset = XML_CHAR_ENCODING_UTF8; |
1295 | } else |
1296 | return(-1); |
1297 | return(ret); |
1298 | } |
1299 | |
1300 | /** |
1301 | * xmlSwitchToEncoding: |
1302 | * @ctxt: the parser context |
1303 | * @handler: the encoding handler |
1304 | * |
1305 | * change the input functions when discovering the character encoding |
1306 | * of a given entity. |
1307 | * |
1308 | * Returns 0 in case of success, -1 otherwise |
1309 | */ |
1310 | int |
1311 | xmlSwitchToEncoding(xmlParserCtxtPtr ctxt, xmlCharEncodingHandlerPtr handler) |
1312 | { |
1313 | return (xmlSwitchToEncodingInt(ctxt, handler, -1)); |
1314 | } |
1315 | |
1316 | /************************************************************************ |
1317 | * * |
1318 | * Commodity functions to handle entities processing * |
1319 | * * |
1320 | ************************************************************************/ |
1321 | |
1322 | /** |
1323 | * xmlFreeInputStream: |
1324 | * @input: an xmlParserInputPtr |
1325 | * |
1326 | * Free up an input stream. |
1327 | */ |
1328 | void |
1329 | xmlFreeInputStream(xmlParserInputPtr input) { |
1330 | if (input == NULL) return; |
1331 | |
1332 | if (input->filename != NULL) xmlFree((char *) input->filename); |
1333 | if (input->directory != NULL) xmlFree((char *) input->directory); |
1334 | if (input->encoding != NULL) xmlFree((char *) input->encoding); |
1335 | if (input->version != NULL) xmlFree((char *) input->version); |
1336 | if ((input->free != NULL) && (input->base != NULL)) |
1337 | input->free((xmlChar *) input->base); |
1338 | if (input->buf != NULL) |
1339 | xmlFreeParserInputBuffer(input->buf); |
1340 | xmlFree(input); |
1341 | } |
1342 | |
1343 | /** |
1344 | * xmlNewInputStream: |
1345 | * @ctxt: an XML parser context |
1346 | * |
1347 | * Create a new input stream structure. |
1348 | * |
1349 | * Returns the new input stream or NULL |
1350 | */ |
1351 | xmlParserInputPtr |
1352 | xmlNewInputStream(xmlParserCtxtPtr ctxt) { |
1353 | xmlParserInputPtr input; |
1354 | |
1355 | input = (xmlParserInputPtr) xmlMalloc(sizeof(xmlParserInput)); |
1356 | if (input == NULL) { |
1357 | xmlErrMemory(ctxt, "couldn't allocate a new input stream\n" ); |
1358 | return(NULL); |
1359 | } |
1360 | memset(input, 0, sizeof(xmlParserInput)); |
1361 | input->line = 1; |
1362 | input->col = 1; |
1363 | input->standalone = -1; |
1364 | |
1365 | /* |
1366 | * If the context is NULL the id cannot be initialized, but that |
1367 | * should not happen while parsing which is the situation where |
1368 | * the id is actually needed. |
1369 | */ |
1370 | if (ctxt != NULL) |
1371 | input->id = ctxt->input_id++; |
1372 | |
1373 | return(input); |
1374 | } |
1375 | |
1376 | /** |
1377 | * xmlNewIOInputStream: |
1378 | * @ctxt: an XML parser context |
1379 | * @input: an I/O Input |
1380 | * @enc: the charset encoding if known |
1381 | * |
1382 | * Create a new input stream structure encapsulating the @input into |
1383 | * a stream suitable for the parser. |
1384 | * |
1385 | * Returns the new input stream or NULL |
1386 | */ |
1387 | xmlParserInputPtr |
1388 | xmlNewIOInputStream(xmlParserCtxtPtr ctxt, xmlParserInputBufferPtr input, |
1389 | xmlCharEncoding enc) { |
1390 | xmlParserInputPtr inputStream; |
1391 | |
1392 | if (input == NULL) return(NULL); |
1393 | if (xmlParserDebugEntities) |
1394 | xmlGenericError(xmlGenericErrorContext, "new input from I/O\n" ); |
1395 | inputStream = xmlNewInputStream(ctxt); |
1396 | if (inputStream == NULL) { |
1397 | return(NULL); |
1398 | } |
1399 | inputStream->filename = NULL; |
1400 | inputStream->buf = input; |
1401 | xmlBufResetInput(inputStream->buf->buffer, inputStream); |
1402 | |
1403 | if (enc != XML_CHAR_ENCODING_NONE) { |
1404 | xmlSwitchEncoding(ctxt, enc); |
1405 | } |
1406 | |
1407 | return(inputStream); |
1408 | } |
1409 | |
1410 | /** |
1411 | * xmlNewEntityInputStream: |
1412 | * @ctxt: an XML parser context |
1413 | * @entity: an Entity pointer |
1414 | * |
1415 | * Create a new input stream based on an xmlEntityPtr |
1416 | * |
1417 | * Returns the new input stream or NULL |
1418 | */ |
1419 | xmlParserInputPtr |
1420 | xmlNewEntityInputStream(xmlParserCtxtPtr ctxt, xmlEntityPtr entity) { |
1421 | xmlParserInputPtr input; |
1422 | |
1423 | if (entity == NULL) { |
1424 | xmlErrInternal(ctxt, "xmlNewEntityInputStream entity = NULL\n" , |
1425 | NULL); |
1426 | return(NULL); |
1427 | } |
1428 | if (xmlParserDebugEntities) |
1429 | xmlGenericError(xmlGenericErrorContext, |
1430 | "new input from entity: %s\n" , entity->name); |
1431 | if (entity->content == NULL) { |
1432 | switch (entity->etype) { |
1433 | case XML_EXTERNAL_GENERAL_UNPARSED_ENTITY: |
1434 | xmlErrInternal(ctxt, "Cannot parse entity %s\n" , |
1435 | entity->name); |
1436 | break; |
1437 | case XML_EXTERNAL_GENERAL_PARSED_ENTITY: |
1438 | case XML_EXTERNAL_PARAMETER_ENTITY: |
1439 | return(xmlLoadExternalEntity((char *) entity->URI, |
1440 | (char *) entity->ExternalID, ctxt)); |
1441 | case XML_INTERNAL_GENERAL_ENTITY: |
1442 | xmlErrInternal(ctxt, |
1443 | "Internal entity %s without content !\n" , |
1444 | entity->name); |
1445 | break; |
1446 | case XML_INTERNAL_PARAMETER_ENTITY: |
1447 | xmlErrInternal(ctxt, |
1448 | "Internal parameter entity %s without content !\n" , |
1449 | entity->name); |
1450 | break; |
1451 | case XML_INTERNAL_PREDEFINED_ENTITY: |
1452 | xmlErrInternal(ctxt, |
1453 | "Predefined entity %s without content !\n" , |
1454 | entity->name); |
1455 | break; |
1456 | } |
1457 | return(NULL); |
1458 | } |
1459 | input = xmlNewInputStream(ctxt); |
1460 | if (input == NULL) { |
1461 | return(NULL); |
1462 | } |
1463 | if (entity->URI != NULL) |
1464 | input->filename = (char *) xmlStrdup((xmlChar *) entity->URI); |
1465 | input->base = entity->content; |
1466 | if (entity->length == 0) |
1467 | entity->length = xmlStrlen(entity->content); |
1468 | input->cur = entity->content; |
1469 | input->length = entity->length; |
1470 | input->end = &entity->content[input->length]; |
1471 | return(input); |
1472 | } |
1473 | |
1474 | /** |
1475 | * xmlNewStringInputStream: |
1476 | * @ctxt: an XML parser context |
1477 | * @buffer: an memory buffer |
1478 | * |
1479 | * Create a new input stream based on a memory buffer. |
1480 | * Returns the new input stream |
1481 | */ |
1482 | xmlParserInputPtr |
1483 | xmlNewStringInputStream(xmlParserCtxtPtr ctxt, const xmlChar *buffer) { |
1484 | xmlParserInputPtr input; |
1485 | |
1486 | if (buffer == NULL) { |
1487 | xmlErrInternal(ctxt, "xmlNewStringInputStream string = NULL\n" , |
1488 | NULL); |
1489 | return(NULL); |
1490 | } |
1491 | if (xmlParserDebugEntities) |
1492 | xmlGenericError(xmlGenericErrorContext, |
1493 | "new fixed input: %.30s\n" , buffer); |
1494 | input = xmlNewInputStream(ctxt); |
1495 | if (input == NULL) { |
1496 | xmlErrMemory(ctxt, "couldn't allocate a new input stream\n" ); |
1497 | return(NULL); |
1498 | } |
1499 | input->base = buffer; |
1500 | input->cur = buffer; |
1501 | input->length = xmlStrlen(buffer); |
1502 | input->end = &buffer[input->length]; |
1503 | return(input); |
1504 | } |
1505 | |
1506 | /** |
1507 | * xmlNewInputFromFile: |
1508 | * @ctxt: an XML parser context |
1509 | * @filename: the filename to use as entity |
1510 | * |
1511 | * Create a new input stream based on a file or an URL. |
1512 | * |
1513 | * Returns the new input stream or NULL in case of error |
1514 | */ |
1515 | xmlParserInputPtr |
1516 | xmlNewInputFromFile(xmlParserCtxtPtr ctxt, const char *filename) { |
1517 | xmlParserInputBufferPtr buf; |
1518 | xmlParserInputPtr inputStream; |
1519 | char *directory = NULL; |
1520 | xmlChar *URI = NULL; |
1521 | |
1522 | if (xmlParserDebugEntities) |
1523 | xmlGenericError(xmlGenericErrorContext, |
1524 | "new input from file: %s\n" , filename); |
1525 | if (ctxt == NULL) return(NULL); |
1526 | buf = xmlParserInputBufferCreateFilename(filename, XML_CHAR_ENCODING_NONE); |
1527 | if (buf == NULL) { |
1528 | if (filename == NULL) |
1529 | __xmlLoaderErr(ctxt, |
1530 | "failed to load external entity: NULL filename \n" , |
1531 | NULL); |
1532 | else |
1533 | __xmlLoaderErr(ctxt, "failed to load external entity \"%s\"\n" , |
1534 | (const char *) filename); |
1535 | return(NULL); |
1536 | } |
1537 | |
1538 | inputStream = xmlNewInputStream(ctxt); |
1539 | if (inputStream == NULL) |
1540 | return(NULL); |
1541 | |
1542 | inputStream->buf = buf; |
1543 | inputStream = xmlCheckHTTPInput(ctxt, inputStream); |
1544 | if (inputStream == NULL) |
1545 | return(NULL); |
1546 | |
1547 | if (inputStream->filename == NULL) |
1548 | URI = xmlStrdup((xmlChar *) filename); |
1549 | else |
1550 | URI = xmlStrdup((xmlChar *) inputStream->filename); |
1551 | directory = xmlParserGetDirectory((const char *) URI); |
1552 | if (inputStream->filename != NULL) xmlFree((char *)inputStream->filename); |
1553 | inputStream->filename = (char *) xmlCanonicPath((const xmlChar *) URI); |
1554 | if (URI != NULL) xmlFree((char *) URI); |
1555 | inputStream->directory = directory; |
1556 | |
1557 | xmlBufResetInput(inputStream->buf->buffer, inputStream); |
1558 | if ((ctxt->directory == NULL) && (directory != NULL)) |
1559 | ctxt->directory = (char *) xmlStrdup((const xmlChar *) directory); |
1560 | return(inputStream); |
1561 | } |
1562 | |
1563 | /************************************************************************ |
1564 | * * |
1565 | * Commodity functions to handle parser contexts * |
1566 | * * |
1567 | ************************************************************************/ |
1568 | |
1569 | /** |
1570 | * xmlInitParserCtxt: |
1571 | * @ctxt: an XML parser context |
1572 | * |
1573 | * Initialize a parser context |
1574 | * |
1575 | * Returns 0 in case of success and -1 in case of error |
1576 | */ |
1577 | |
1578 | int |
1579 | xmlInitParserCtxt(xmlParserCtxtPtr ctxt) |
1580 | { |
1581 | xmlParserInputPtr input; |
1582 | |
1583 | if(ctxt==NULL) { |
1584 | xmlErrInternal(NULL, "Got NULL parser context\n" , NULL); |
1585 | return(-1); |
1586 | } |
1587 | |
1588 | xmlDefaultSAXHandlerInit(); |
1589 | |
1590 | if (ctxt->dict == NULL) |
1591 | ctxt->dict = xmlDictCreate(); |
1592 | if (ctxt->dict == NULL) { |
1593 | xmlErrMemory(NULL, "cannot initialize parser context\n" ); |
1594 | return(-1); |
1595 | } |
1596 | xmlDictSetLimit(ctxt->dict, XML_MAX_DICTIONARY_LIMIT); |
1597 | |
1598 | if (ctxt->sax == NULL) |
1599 | ctxt->sax = (xmlSAXHandler *) xmlMalloc(sizeof(xmlSAXHandler)); |
1600 | if (ctxt->sax == NULL) { |
1601 | xmlErrMemory(NULL, "cannot initialize parser context\n" ); |
1602 | return(-1); |
1603 | } |
1604 | else |
1605 | xmlSAXVersion(ctxt->sax, 2); |
1606 | |
1607 | ctxt->maxatts = 0; |
1608 | ctxt->atts = NULL; |
1609 | /* Allocate the Input stack */ |
1610 | if (ctxt->inputTab == NULL) { |
1611 | ctxt->inputTab = (xmlParserInputPtr *) |
1612 | xmlMalloc(5 * sizeof(xmlParserInputPtr)); |
1613 | ctxt->inputMax = 5; |
1614 | } |
1615 | if (ctxt->inputTab == NULL) { |
1616 | xmlErrMemory(NULL, "cannot initialize parser context\n" ); |
1617 | ctxt->inputNr = 0; |
1618 | ctxt->inputMax = 0; |
1619 | ctxt->input = NULL; |
1620 | return(-1); |
1621 | } |
1622 | while ((input = inputPop(ctxt)) != NULL) { /* Non consuming */ |
1623 | xmlFreeInputStream(input); |
1624 | } |
1625 | ctxt->inputNr = 0; |
1626 | ctxt->input = NULL; |
1627 | |
1628 | ctxt->version = NULL; |
1629 | ctxt->encoding = NULL; |
1630 | ctxt->standalone = -1; |
1631 | ctxt->hasExternalSubset = 0; |
1632 | ctxt->hasPErefs = 0; |
1633 | ctxt->html = 0; |
1634 | ctxt->external = 0; |
1635 | ctxt->instate = XML_PARSER_START; |
1636 | ctxt->token = 0; |
1637 | ctxt->directory = NULL; |
1638 | |
1639 | /* Allocate the Node stack */ |
1640 | if (ctxt->nodeTab == NULL) { |
1641 | ctxt->nodeTab = (xmlNodePtr *) xmlMalloc(10 * sizeof(xmlNodePtr)); |
1642 | ctxt->nodeMax = 10; |
1643 | } |
1644 | if (ctxt->nodeTab == NULL) { |
1645 | xmlErrMemory(NULL, "cannot initialize parser context\n" ); |
1646 | ctxt->nodeNr = 0; |
1647 | ctxt->nodeMax = 0; |
1648 | ctxt->node = NULL; |
1649 | ctxt->inputNr = 0; |
1650 | ctxt->inputMax = 0; |
1651 | ctxt->input = NULL; |
1652 | return(-1); |
1653 | } |
1654 | ctxt->nodeNr = 0; |
1655 | ctxt->node = NULL; |
1656 | |
1657 | /* Allocate the Name stack */ |
1658 | if (ctxt->nameTab == NULL) { |
1659 | ctxt->nameTab = (const xmlChar **) xmlMalloc(10 * sizeof(xmlChar *)); |
1660 | ctxt->nameMax = 10; |
1661 | } |
1662 | if (ctxt->nameTab == NULL) { |
1663 | xmlErrMemory(NULL, "cannot initialize parser context\n" ); |
1664 | ctxt->nodeNr = 0; |
1665 | ctxt->nodeMax = 0; |
1666 | ctxt->node = NULL; |
1667 | ctxt->inputNr = 0; |
1668 | ctxt->inputMax = 0; |
1669 | ctxt->input = NULL; |
1670 | ctxt->nameNr = 0; |
1671 | ctxt->nameMax = 0; |
1672 | ctxt->name = NULL; |
1673 | return(-1); |
1674 | } |
1675 | ctxt->nameNr = 0; |
1676 | ctxt->name = NULL; |
1677 | |
1678 | /* Allocate the space stack */ |
1679 | if (ctxt->spaceTab == NULL) { |
1680 | ctxt->spaceTab = (int *) xmlMalloc(10 * sizeof(int)); |
1681 | ctxt->spaceMax = 10; |
1682 | } |
1683 | if (ctxt->spaceTab == NULL) { |
1684 | xmlErrMemory(NULL, "cannot initialize parser context\n" ); |
1685 | ctxt->nodeNr = 0; |
1686 | ctxt->nodeMax = 0; |
1687 | ctxt->node = NULL; |
1688 | ctxt->inputNr = 0; |
1689 | ctxt->inputMax = 0; |
1690 | ctxt->input = NULL; |
1691 | ctxt->nameNr = 0; |
1692 | ctxt->nameMax = 0; |
1693 | ctxt->name = NULL; |
1694 | ctxt->spaceNr = 0; |
1695 | ctxt->spaceMax = 0; |
1696 | ctxt->space = NULL; |
1697 | return(-1); |
1698 | } |
1699 | ctxt->spaceNr = 1; |
1700 | ctxt->spaceMax = 10; |
1701 | ctxt->spaceTab[0] = -1; |
1702 | ctxt->space = &ctxt->spaceTab[0]; |
1703 | ctxt->userData = ctxt; |
1704 | ctxt->myDoc = NULL; |
1705 | ctxt->wellFormed = 1; |
1706 | ctxt->nsWellFormed = 1; |
1707 | ctxt->valid = 1; |
1708 | ctxt->loadsubset = xmlLoadExtDtdDefaultValue; |
1709 | if (ctxt->loadsubset) { |
1710 | ctxt->options |= XML_PARSE_DTDLOAD; |
1711 | } |
1712 | ctxt->validate = xmlDoValidityCheckingDefaultValue; |
1713 | ctxt->pedantic = xmlPedanticParserDefaultValue; |
1714 | if (ctxt->pedantic) { |
1715 | ctxt->options |= XML_PARSE_PEDANTIC; |
1716 | } |
1717 | ctxt->linenumbers = xmlLineNumbersDefaultValue; |
1718 | ctxt->keepBlanks = xmlKeepBlanksDefaultValue; |
1719 | if (ctxt->keepBlanks == 0) { |
1720 | ctxt->sax->ignorableWhitespace = xmlSAX2IgnorableWhitespace; |
1721 | ctxt->options |= XML_PARSE_NOBLANKS; |
1722 | } |
1723 | |
1724 | ctxt->vctxt.finishDtd = XML_CTXT_FINISH_DTD_0; |
1725 | ctxt->vctxt.userData = ctxt; |
1726 | ctxt->vctxt.error = xmlParserValidityError; |
1727 | ctxt->vctxt.warning = xmlParserValidityWarning; |
1728 | if (ctxt->validate) { |
1729 | if (xmlGetWarningsDefaultValue == 0) |
1730 | ctxt->vctxt.warning = NULL; |
1731 | else |
1732 | ctxt->vctxt.warning = xmlParserValidityWarning; |
1733 | ctxt->vctxt.nodeMax = 0; |
1734 | ctxt->options |= XML_PARSE_DTDVALID; |
1735 | } |
1736 | ctxt->replaceEntities = xmlSubstituteEntitiesDefaultValue; |
1737 | if (ctxt->replaceEntities) { |
1738 | ctxt->options |= XML_PARSE_NOENT; |
1739 | } |
1740 | ctxt->record_info = 0; |
1741 | ctxt->nbChars = 0; |
1742 | ctxt->checkIndex = 0; |
1743 | ctxt->inSubset = 0; |
1744 | ctxt->errNo = XML_ERR_OK; |
1745 | ctxt->depth = 0; |
1746 | ctxt->charset = XML_CHAR_ENCODING_UTF8; |
1747 | ctxt->catalogs = NULL; |
1748 | ctxt->nbentities = 0; |
1749 | ctxt->sizeentities = 0; |
1750 | ctxt->sizeentcopy = 0; |
1751 | ctxt->input_id = 1; |
1752 | xmlInitNodeInfoSeq(&ctxt->node_seq); |
1753 | return(0); |
1754 | } |
1755 | |
1756 | /** |
1757 | * xmlFreeParserCtxt: |
1758 | * @ctxt: an XML parser context |
1759 | * |
1760 | * Free all the memory used by a parser context. However the parsed |
1761 | * document in ctxt->myDoc is not freed. |
1762 | */ |
1763 | |
1764 | void |
1765 | xmlFreeParserCtxt(xmlParserCtxtPtr ctxt) |
1766 | { |
1767 | xmlParserInputPtr input; |
1768 | |
1769 | if (ctxt == NULL) return; |
1770 | |
1771 | while ((input = inputPop(ctxt)) != NULL) { /* Non consuming */ |
1772 | xmlFreeInputStream(input); |
1773 | } |
1774 | if (ctxt->spaceTab != NULL) xmlFree(ctxt->spaceTab); |
1775 | if (ctxt->nameTab != NULL) xmlFree((xmlChar * *)ctxt->nameTab); |
1776 | if (ctxt->nodeTab != NULL) xmlFree(ctxt->nodeTab); |
1777 | if (ctxt->nodeInfoTab != NULL) xmlFree(ctxt->nodeInfoTab); |
1778 | if (ctxt->inputTab != NULL) xmlFree(ctxt->inputTab); |
1779 | if (ctxt->version != NULL) xmlFree((char *) ctxt->version); |
1780 | if (ctxt->encoding != NULL) xmlFree((char *) ctxt->encoding); |
1781 | if (ctxt->extSubURI != NULL) xmlFree((char *) ctxt->extSubURI); |
1782 | if (ctxt->extSubSystem != NULL) xmlFree((char *) ctxt->extSubSystem); |
1783 | #ifdef LIBXML_SAX1_ENABLED |
1784 | if ((ctxt->sax != NULL) && |
1785 | (ctxt->sax != (xmlSAXHandlerPtr) &xmlDefaultSAXHandler)) |
1786 | #else |
1787 | if (ctxt->sax != NULL) |
1788 | #endif /* LIBXML_SAX1_ENABLED */ |
1789 | xmlFree(ctxt->sax); |
1790 | if (ctxt->directory != NULL) xmlFree((char *) ctxt->directory); |
1791 | if (ctxt->vctxt.nodeTab != NULL) xmlFree(ctxt->vctxt.nodeTab); |
1792 | if (ctxt->atts != NULL) xmlFree((xmlChar * *)ctxt->atts); |
1793 | if (ctxt->dict != NULL) xmlDictFree(ctxt->dict); |
1794 | if (ctxt->nsTab != NULL) xmlFree((char *) ctxt->nsTab); |
1795 | if (ctxt->pushTab != NULL) xmlFree(ctxt->pushTab); |
1796 | if (ctxt->attallocs != NULL) xmlFree(ctxt->attallocs); |
1797 | if (ctxt->attsDefault != NULL) |
1798 | xmlHashFree(ctxt->attsDefault, xmlHashDefaultDeallocator); |
1799 | if (ctxt->attsSpecial != NULL) |
1800 | xmlHashFree(ctxt->attsSpecial, NULL); |
1801 | if (ctxt->freeElems != NULL) { |
1802 | xmlNodePtr cur, next; |
1803 | |
1804 | cur = ctxt->freeElems; |
1805 | while (cur != NULL) { |
1806 | next = cur->next; |
1807 | xmlFree(cur); |
1808 | cur = next; |
1809 | } |
1810 | } |
1811 | if (ctxt->freeAttrs != NULL) { |
1812 | xmlAttrPtr cur, next; |
1813 | |
1814 | cur = ctxt->freeAttrs; |
1815 | while (cur != NULL) { |
1816 | next = cur->next; |
1817 | xmlFree(cur); |
1818 | cur = next; |
1819 | } |
1820 | } |
1821 | /* |
1822 | * cleanup the error strings |
1823 | */ |
1824 | if (ctxt->lastError.message != NULL) |
1825 | xmlFree(ctxt->lastError.message); |
1826 | if (ctxt->lastError.file != NULL) |
1827 | xmlFree(ctxt->lastError.file); |
1828 | if (ctxt->lastError.str1 != NULL) |
1829 | xmlFree(ctxt->lastError.str1); |
1830 | if (ctxt->lastError.str2 != NULL) |
1831 | xmlFree(ctxt->lastError.str2); |
1832 | if (ctxt->lastError.str3 != NULL) |
1833 | xmlFree(ctxt->lastError.str3); |
1834 | |
1835 | #ifdef LIBXML_CATALOG_ENABLED |
1836 | if (ctxt->catalogs != NULL) |
1837 | xmlCatalogFreeLocal(ctxt->catalogs); |
1838 | #endif |
1839 | xmlFree(ctxt); |
1840 | } |
1841 | |
1842 | /** |
1843 | * xmlNewParserCtxt: |
1844 | * |
1845 | * Allocate and initialize a new parser context. |
1846 | * |
1847 | * Returns the xmlParserCtxtPtr or NULL |
1848 | */ |
1849 | |
1850 | xmlParserCtxtPtr |
1851 | xmlNewParserCtxt(void) |
1852 | { |
1853 | xmlParserCtxtPtr ctxt; |
1854 | |
1855 | ctxt = (xmlParserCtxtPtr) xmlMalloc(sizeof(xmlParserCtxt)); |
1856 | if (ctxt == NULL) { |
1857 | xmlErrMemory(NULL, "cannot allocate parser context\n" ); |
1858 | return(NULL); |
1859 | } |
1860 | memset(ctxt, 0, sizeof(xmlParserCtxt)); |
1861 | if (xmlInitParserCtxt(ctxt) < 0) { |
1862 | xmlFreeParserCtxt(ctxt); |
1863 | return(NULL); |
1864 | } |
1865 | return(ctxt); |
1866 | } |
1867 | |
1868 | /************************************************************************ |
1869 | * * |
1870 | * Handling of node informations * |
1871 | * * |
1872 | ************************************************************************/ |
1873 | |
1874 | /** |
1875 | * xmlClearParserCtxt: |
1876 | * @ctxt: an XML parser context |
1877 | * |
1878 | * Clear (release owned resources) and reinitialize a parser context |
1879 | */ |
1880 | |
1881 | void |
1882 | xmlClearParserCtxt(xmlParserCtxtPtr ctxt) |
1883 | { |
1884 | if (ctxt==NULL) |
1885 | return; |
1886 | xmlClearNodeInfoSeq(&ctxt->node_seq); |
1887 | xmlCtxtReset(ctxt); |
1888 | } |
1889 | |
1890 | |
1891 | /** |
1892 | * xmlParserFindNodeInfo: |
1893 | * @ctx: an XML parser context |
1894 | * @node: an XML node within the tree |
1895 | * |
1896 | * Find the parser node info struct for a given node |
1897 | * |
1898 | * Returns an xmlParserNodeInfo block pointer or NULL |
1899 | */ |
1900 | const xmlParserNodeInfo * |
1901 | xmlParserFindNodeInfo(const xmlParserCtxtPtr ctx, const xmlNodePtr node) |
1902 | { |
1903 | unsigned long pos; |
1904 | |
1905 | if ((ctx == NULL) || (node == NULL)) |
1906 | return (NULL); |
1907 | /* Find position where node should be at */ |
1908 | pos = xmlParserFindNodeInfoIndex(&ctx->node_seq, node); |
1909 | if (pos < ctx->node_seq.length |
1910 | && ctx->node_seq.buffer[pos].node == node) |
1911 | return &ctx->node_seq.buffer[pos]; |
1912 | else |
1913 | return NULL; |
1914 | } |
1915 | |
1916 | |
1917 | /** |
1918 | * xmlInitNodeInfoSeq: |
1919 | * @seq: a node info sequence pointer |
1920 | * |
1921 | * -- Initialize (set to initial state) node info sequence |
1922 | */ |
1923 | void |
1924 | xmlInitNodeInfoSeq(xmlParserNodeInfoSeqPtr seq) |
1925 | { |
1926 | if (seq == NULL) |
1927 | return; |
1928 | seq->length = 0; |
1929 | seq->maximum = 0; |
1930 | seq->buffer = NULL; |
1931 | } |
1932 | |
1933 | /** |
1934 | * xmlClearNodeInfoSeq: |
1935 | * @seq: a node info sequence pointer |
1936 | * |
1937 | * -- Clear (release memory and reinitialize) node |
1938 | * info sequence |
1939 | */ |
1940 | void |
1941 | xmlClearNodeInfoSeq(xmlParserNodeInfoSeqPtr seq) |
1942 | { |
1943 | if (seq == NULL) |
1944 | return; |
1945 | if (seq->buffer != NULL) |
1946 | xmlFree(seq->buffer); |
1947 | xmlInitNodeInfoSeq(seq); |
1948 | } |
1949 | |
1950 | /** |
1951 | * xmlParserFindNodeInfoIndex: |
1952 | * @seq: a node info sequence pointer |
1953 | * @node: an XML node pointer |
1954 | * |
1955 | * |
1956 | * xmlParserFindNodeInfoIndex : Find the index that the info record for |
1957 | * the given node is or should be at in a sorted sequence |
1958 | * |
1959 | * Returns a long indicating the position of the record |
1960 | */ |
1961 | unsigned long |
1962 | xmlParserFindNodeInfoIndex(const xmlParserNodeInfoSeqPtr seq, |
1963 | const xmlNodePtr node) |
1964 | { |
1965 | unsigned long upper, lower, middle; |
1966 | int found = 0; |
1967 | |
1968 | if ((seq == NULL) || (node == NULL)) |
1969 | return ((unsigned long) -1); |
1970 | |
1971 | /* Do a binary search for the key */ |
1972 | lower = 1; |
1973 | upper = seq->length; |
1974 | middle = 0; |
1975 | while (lower <= upper && !found) { |
1976 | middle = lower + (upper - lower) / 2; |
1977 | if (node == seq->buffer[middle - 1].node) |
1978 | found = 1; |
1979 | else if (node < seq->buffer[middle - 1].node) |
1980 | upper = middle - 1; |
1981 | else |
1982 | lower = middle + 1; |
1983 | } |
1984 | |
1985 | /* Return position */ |
1986 | if (middle == 0 || seq->buffer[middle - 1].node < node) |
1987 | return middle; |
1988 | else |
1989 | return middle - 1; |
1990 | } |
1991 | |
1992 | |
1993 | /** |
1994 | * xmlParserAddNodeInfo: |
1995 | * @ctxt: an XML parser context |
1996 | * @info: a node info sequence pointer |
1997 | * |
1998 | * Insert node info record into the sorted sequence |
1999 | */ |
2000 | void |
2001 | xmlParserAddNodeInfo(xmlParserCtxtPtr ctxt, |
2002 | const xmlParserNodeInfoPtr info) |
2003 | { |
2004 | unsigned long pos; |
2005 | |
2006 | if ((ctxt == NULL) || (info == NULL)) return; |
2007 | |
2008 | /* Find pos and check to see if node is already in the sequence */ |
2009 | pos = xmlParserFindNodeInfoIndex(&ctxt->node_seq, (xmlNodePtr) |
2010 | info->node); |
2011 | |
2012 | if ((pos < ctxt->node_seq.length) && |
2013 | (ctxt->node_seq.buffer != NULL) && |
2014 | (ctxt->node_seq.buffer[pos].node == info->node)) { |
2015 | ctxt->node_seq.buffer[pos] = *info; |
2016 | } |
2017 | |
2018 | /* Otherwise, we need to add new node to buffer */ |
2019 | else { |
2020 | if ((ctxt->node_seq.length + 1 > ctxt->node_seq.maximum) || |
2021 | (ctxt->node_seq.buffer == NULL)) { |
2022 | xmlParserNodeInfo *tmp_buffer; |
2023 | unsigned int byte_size; |
2024 | |
2025 | if (ctxt->node_seq.maximum == 0) |
2026 | ctxt->node_seq.maximum = 2; |
2027 | byte_size = (sizeof(*ctxt->node_seq.buffer) * |
2028 | (2 * ctxt->node_seq.maximum)); |
2029 | |
2030 | if (ctxt->node_seq.buffer == NULL) |
2031 | tmp_buffer = (xmlParserNodeInfo *) xmlMalloc(byte_size); |
2032 | else |
2033 | tmp_buffer = |
2034 | (xmlParserNodeInfo *) xmlRealloc(ctxt->node_seq.buffer, |
2035 | byte_size); |
2036 | |
2037 | if (tmp_buffer == NULL) { |
2038 | xmlErrMemory(ctxt, "failed to allocate buffer\n" ); |
2039 | return; |
2040 | } |
2041 | ctxt->node_seq.buffer = tmp_buffer; |
2042 | ctxt->node_seq.maximum *= 2; |
2043 | } |
2044 | |
2045 | /* If position is not at end, move elements out of the way */ |
2046 | if (pos != ctxt->node_seq.length) { |
2047 | unsigned long i; |
2048 | |
2049 | for (i = ctxt->node_seq.length; i > pos; i--) |
2050 | ctxt->node_seq.buffer[i] = ctxt->node_seq.buffer[i - 1]; |
2051 | } |
2052 | |
2053 | /* Copy element and increase length */ |
2054 | ctxt->node_seq.buffer[pos] = *info; |
2055 | ctxt->node_seq.length++; |
2056 | } |
2057 | } |
2058 | |
2059 | /************************************************************************ |
2060 | * * |
2061 | * Defaults settings * |
2062 | * * |
2063 | ************************************************************************/ |
2064 | /** |
2065 | * xmlPedanticParserDefault: |
2066 | * @val: int 0 or 1 |
2067 | * |
2068 | * Set and return the previous value for enabling pedantic warnings. |
2069 | * |
2070 | * Returns the last value for 0 for no substitution, 1 for substitution. |
2071 | */ |
2072 | |
2073 | int |
2074 | xmlPedanticParserDefault(int val) { |
2075 | int old = xmlPedanticParserDefaultValue; |
2076 | |
2077 | xmlPedanticParserDefaultValue = val; |
2078 | return(old); |
2079 | } |
2080 | |
2081 | /** |
2082 | * xmlLineNumbersDefault: |
2083 | * @val: int 0 or 1 |
2084 | * |
2085 | * Set and return the previous value for enabling line numbers in elements |
2086 | * contents. This may break on old application and is turned off by default. |
2087 | * |
2088 | * Returns the last value for 0 for no substitution, 1 for substitution. |
2089 | */ |
2090 | |
2091 | int |
2092 | xmlLineNumbersDefault(int val) { |
2093 | int old = xmlLineNumbersDefaultValue; |
2094 | |
2095 | xmlLineNumbersDefaultValue = val; |
2096 | return(old); |
2097 | } |
2098 | |
2099 | /** |
2100 | * xmlSubstituteEntitiesDefault: |
2101 | * @val: int 0 or 1 |
2102 | * |
2103 | * Set and return the previous value for default entity support. |
2104 | * Initially the parser always keep entity references instead of substituting |
2105 | * entity values in the output. This function has to be used to change the |
2106 | * default parser behavior |
2107 | * SAX::substituteEntities() has to be used for changing that on a file by |
2108 | * file basis. |
2109 | * |
2110 | * Returns the last value for 0 for no substitution, 1 for substitution. |
2111 | */ |
2112 | |
2113 | int |
2114 | xmlSubstituteEntitiesDefault(int val) { |
2115 | int old = xmlSubstituteEntitiesDefaultValue; |
2116 | |
2117 | xmlSubstituteEntitiesDefaultValue = val; |
2118 | return(old); |
2119 | } |
2120 | |
2121 | /** |
2122 | * xmlKeepBlanksDefault: |
2123 | * @val: int 0 or 1 |
2124 | * |
2125 | * Set and return the previous value for default blanks text nodes support. |
2126 | * The 1.x version of the parser used an heuristic to try to detect |
2127 | * ignorable white spaces. As a result the SAX callback was generating |
2128 | * xmlSAX2IgnorableWhitespace() callbacks instead of characters() one, and when |
2129 | * using the DOM output text nodes containing those blanks were not generated. |
2130 | * The 2.x and later version will switch to the XML standard way and |
2131 | * ignorableWhitespace() are only generated when running the parser in |
2132 | * validating mode and when the current element doesn't allow CDATA or |
2133 | * mixed content. |
2134 | * This function is provided as a way to force the standard behavior |
2135 | * on 1.X libs and to switch back to the old mode for compatibility when |
2136 | * running 1.X client code on 2.X . Upgrade of 1.X code should be done |
2137 | * by using xmlIsBlankNode() commodity function to detect the "empty" |
2138 | * nodes generated. |
2139 | * This value also affect autogeneration of indentation when saving code |
2140 | * if blanks sections are kept, indentation is not generated. |
2141 | * |
2142 | * Returns the last value for 0 for no substitution, 1 for substitution. |
2143 | */ |
2144 | |
2145 | int |
2146 | xmlKeepBlanksDefault(int val) { |
2147 | int old = xmlKeepBlanksDefaultValue; |
2148 | |
2149 | xmlKeepBlanksDefaultValue = val; |
2150 | if (!val) xmlIndentTreeOutput = 1; |
2151 | return(old); |
2152 | } |
2153 | |
2154 | #define bottom_parserInternals |
2155 | #include "elfgcchack.h" |
2156 | |