Annotation of embedaddon/libxml2/parserInternals.c, revision 1.1.1.2
1.1 misho 1: /*
2: * parserInternals.c : Internal routines (and obsolete ones) needed for the
3: * XML and HTML parsers.
4: *
5: * See Copyright for the status of this software.
6: *
7: * daniel@veillard.com
8: */
9:
10: #define IN_LIBXML
11: #include "libxml.h"
12:
13: #if defined(WIN32) && !defined (__CYGWIN__)
14: #define XML_DIR_SEP '\\'
15: #else
16: #define XML_DIR_SEP '/'
17: #endif
18:
19: #include <string.h>
20: #ifdef HAVE_CTYPE_H
21: #include <ctype.h>
22: #endif
23: #ifdef HAVE_STDLIB_H
24: #include <stdlib.h>
25: #endif
26: #ifdef HAVE_SYS_STAT_H
27: #include <sys/stat.h>
28: #endif
29: #ifdef HAVE_FCNTL_H
30: #include <fcntl.h>
31: #endif
32: #ifdef HAVE_UNISTD_H
33: #include <unistd.h>
34: #endif
35: #ifdef HAVE_ZLIB_H
36: #include <zlib.h>
37: #endif
38:
39: #include <libxml/xmlmemory.h>
40: #include <libxml/tree.h>
41: #include <libxml/parser.h>
42: #include <libxml/parserInternals.h>
43: #include <libxml/valid.h>
44: #include <libxml/entities.h>
45: #include <libxml/xmlerror.h>
46: #include <libxml/encoding.h>
47: #include <libxml/valid.h>
48: #include <libxml/xmlIO.h>
49: #include <libxml/uri.h>
50: #include <libxml/dict.h>
51: #include <libxml/SAX.h>
52: #ifdef LIBXML_CATALOG_ENABLED
53: #include <libxml/catalog.h>
54: #endif
55: #include <libxml/globals.h>
56: #include <libxml/chvalid.h>
57:
58: /*
59: * Various global defaults for parsing
60: */
61:
62: /**
63: * xmlCheckVersion:
64: * @version: the include version number
65: *
66: * check the compiled lib version against the include one.
67: * This can warn or immediately kill the application
68: */
69: void
70: xmlCheckVersion(int version) {
71: int myversion = (int) LIBXML_VERSION;
72:
73: xmlInitParser();
74:
75: if ((myversion / 10000) != (version / 10000)) {
76: xmlGenericError(xmlGenericErrorContext,
77: "Fatal: program compiled against libxml %d using libxml %d\n",
78: (version / 10000), (myversion / 10000));
79: fprintf(stderr,
80: "Fatal: program compiled against libxml %d using libxml %d\n",
81: (version / 10000), (myversion / 10000));
82: }
83: if ((myversion / 100) < (version / 100)) {
84: xmlGenericError(xmlGenericErrorContext,
85: "Warning: program compiled against libxml %d using older %d\n",
86: (version / 100), (myversion / 100));
87: }
88: }
89:
90:
91: /************************************************************************
92: * *
93: * Some factorized error routines *
94: * *
95: ************************************************************************/
96:
97:
98: /**
99: * xmlErrMemory:
100: * @ctxt: an XML parser context
101: * @extra: extra informations
102: *
103: * Handle a redefinition of attribute error
104: */
105: void
106: xmlErrMemory(xmlParserCtxtPtr ctxt, const char *extra)
107: {
108: if ((ctxt != NULL) && (ctxt->disableSAX != 0) &&
109: (ctxt->instate == XML_PARSER_EOF))
110: return;
111: if (ctxt != NULL) {
112: ctxt->errNo = XML_ERR_NO_MEMORY;
113: ctxt->instate = XML_PARSER_EOF;
114: ctxt->disableSAX = 1;
115: }
116: if (extra)
117: __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER,
118: XML_ERR_NO_MEMORY, XML_ERR_FATAL, NULL, 0, extra,
119: NULL, NULL, 0, 0,
120: "Memory allocation failed : %s\n", extra);
121: else
122: __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER,
123: XML_ERR_NO_MEMORY, XML_ERR_FATAL, NULL, 0, NULL,
124: NULL, NULL, 0, 0, "Memory allocation failed\n");
125: }
126:
127: /**
128: * __xmlErrEncoding:
129: * @ctxt: an XML parser context
130: * @xmlerr: the error number
131: * @msg: the error message
132: * @str1: an string info
133: * @str2: an string info
134: *
135: * Handle an encoding error
136: */
137: void
138: __xmlErrEncoding(xmlParserCtxtPtr ctxt, xmlParserErrors xmlerr,
139: const char *msg, const xmlChar * str1, const xmlChar * str2)
140: {
141: if ((ctxt != NULL) && (ctxt->disableSAX != 0) &&
142: (ctxt->instate == XML_PARSER_EOF))
143: return;
144: if (ctxt != NULL)
145: ctxt->errNo = xmlerr;
146: __xmlRaiseError(NULL, NULL, NULL,
147: ctxt, NULL, XML_FROM_PARSER, xmlerr, XML_ERR_FATAL,
148: NULL, 0, (const char *) str1, (const char *) str2,
149: NULL, 0, 0, msg, str1, str2);
150: if (ctxt != NULL) {
151: ctxt->wellFormed = 0;
152: if (ctxt->recovery == 0)
153: ctxt->disableSAX = 1;
154: }
155: }
156:
157: /**
158: * xmlErrInternal:
159: * @ctxt: an XML parser context
160: * @msg: the error message
161: * @str: error informations
162: *
163: * Handle an internal error
164: */
165: static void
166: xmlErrInternal(xmlParserCtxtPtr ctxt, const char *msg, const xmlChar * str)
167: {
168: if ((ctxt != NULL) && (ctxt->disableSAX != 0) &&
169: (ctxt->instate == XML_PARSER_EOF))
170: return;
171: if (ctxt != NULL)
172: ctxt->errNo = XML_ERR_INTERNAL_ERROR;
173: __xmlRaiseError(NULL, NULL, NULL,
174: ctxt, NULL, XML_FROM_PARSER, XML_ERR_INTERNAL_ERROR,
175: XML_ERR_FATAL, NULL, 0, (const char *) str, NULL, NULL,
176: 0, 0, msg, str);
177: if (ctxt != NULL) {
178: ctxt->wellFormed = 0;
179: if (ctxt->recovery == 0)
180: ctxt->disableSAX = 1;
181: }
182: }
183:
184: /**
185: * xmlErrEncodingInt:
186: * @ctxt: an XML parser context
187: * @error: the error number
188: * @msg: the error message
189: * @val: an integer value
190: *
191: * n encoding error
192: */
193: static void
194: xmlErrEncodingInt(xmlParserCtxtPtr ctxt, xmlParserErrors error,
195: const char *msg, int val)
196: {
197: if ((ctxt != NULL) && (ctxt->disableSAX != 0) &&
198: (ctxt->instate == XML_PARSER_EOF))
199: return;
200: if (ctxt != NULL)
201: ctxt->errNo = error;
202: __xmlRaiseError(NULL, NULL, NULL,
203: ctxt, NULL, XML_FROM_PARSER, error, XML_ERR_FATAL,
204: NULL, 0, NULL, NULL, NULL, val, 0, msg, val);
205: if (ctxt != NULL) {
206: ctxt->wellFormed = 0;
207: if (ctxt->recovery == 0)
208: ctxt->disableSAX = 1;
209: }
210: }
211:
212: /**
213: * xmlIsLetter:
214: * @c: an unicode character (int)
215: *
216: * Check whether the character is allowed by the production
217: * [84] Letter ::= BaseChar | Ideographic
218: *
219: * Returns 0 if not, non-zero otherwise
220: */
221: int
222: xmlIsLetter(int c) {
223: return(IS_BASECHAR(c) || IS_IDEOGRAPHIC(c));
224: }
225:
226: /************************************************************************
227: * *
228: * Input handling functions for progressive parsing *
229: * *
230: ************************************************************************/
231:
232: /* #define DEBUG_INPUT */
233: /* #define DEBUG_STACK */
234: /* #define DEBUG_PUSH */
235:
236:
237: /* we need to keep enough input to show errors in context */
238: #define LINE_LEN 80
239:
240: #ifdef DEBUG_INPUT
241: #define CHECK_BUFFER(in) check_buffer(in)
242:
243: static
244: void check_buffer(xmlParserInputPtr in) {
245: if (in->base != in->buf->buffer->content) {
246: xmlGenericError(xmlGenericErrorContext,
247: "xmlParserInput: base mismatch problem\n");
248: }
249: if (in->cur < in->base) {
250: xmlGenericError(xmlGenericErrorContext,
251: "xmlParserInput: cur < base problem\n");
252: }
253: if (in->cur > in->base + in->buf->buffer->use) {
254: xmlGenericError(xmlGenericErrorContext,
255: "xmlParserInput: cur > base + use problem\n");
256: }
257: xmlGenericError(xmlGenericErrorContext,"buffer %x : content %x, cur %d, use %d, size %d\n",
258: (int) in, (int) in->buf->buffer->content, in->cur - in->base,
259: in->buf->buffer->use, in->buf->buffer->size);
260: }
261:
262: #else
263: #define CHECK_BUFFER(in)
264: #endif
265:
266:
267: /**
268: * xmlParserInputRead:
269: * @in: an XML parser input
270: * @len: an indicative size for the lookahead
271: *
272: * This function refresh the input for the parser. It doesn't try to
273: * preserve pointers to the input buffer, and discard already read data
274: *
275: * Returns the number of xmlChars read, or -1 in case of error, 0 indicate the
276: * end of this entity
277: */
278: int
279: xmlParserInputRead(xmlParserInputPtr in, int len) {
280: int ret;
281: int used;
282: int indx;
283:
284: if (in == NULL) return(-1);
285: #ifdef DEBUG_INPUT
286: xmlGenericError(xmlGenericErrorContext, "Read\n");
287: #endif
288: if (in->buf == NULL) return(-1);
289: if (in->base == NULL) return(-1);
290: if (in->cur == NULL) return(-1);
291: if (in->buf->buffer == NULL) return(-1);
292: if (in->buf->readcallback == NULL) return(-1);
293:
294: CHECK_BUFFER(in);
295:
296: used = in->cur - in->buf->buffer->content;
297: ret = xmlBufferShrink(in->buf->buffer, used);
298: if (ret > 0) {
299: in->cur -= ret;
300: in->consumed += ret;
301: }
302: ret = xmlParserInputBufferRead(in->buf, len);
303: if (in->base != in->buf->buffer->content) {
304: /*
305: * the buffer has been reallocated
306: */
307: indx = in->cur - in->base;
308: in->base = in->buf->buffer->content;
309: in->cur = &in->buf->buffer->content[indx];
310: }
311: in->end = &in->buf->buffer->content[in->buf->buffer->use];
312:
313: CHECK_BUFFER(in);
314:
315: return(ret);
316: }
317:
318: /**
319: * xmlParserInputGrow:
320: * @in: an XML parser input
321: * @len: an indicative size for the lookahead
322: *
323: * This function increase the input for the parser. It tries to
324: * preserve pointers to the input buffer, and keep already read data
325: *
326: * Returns the number of xmlChars read, or -1 in case of error, 0 indicate the
327: * end of this entity
328: */
329: int
330: xmlParserInputGrow(xmlParserInputPtr in, int len) {
331: int ret;
332: int indx;
333:
334: if (in == NULL) return(-1);
335: #ifdef DEBUG_INPUT
336: xmlGenericError(xmlGenericErrorContext, "Grow\n");
337: #endif
338: if (in->buf == NULL) return(-1);
339: if (in->base == NULL) return(-1);
340: if (in->cur == NULL) return(-1);
341: if (in->buf->buffer == NULL) return(-1);
342:
343: CHECK_BUFFER(in);
344:
345: indx = in->cur - in->base;
346: if (in->buf->buffer->use > (unsigned int) indx + INPUT_CHUNK) {
347:
348: CHECK_BUFFER(in);
349:
350: return(0);
351: }
352: if (in->buf->readcallback != NULL)
353: ret = xmlParserInputBufferGrow(in->buf, len);
354: else
355: return(0);
356:
357: /*
358: * NOTE : in->base may be a "dangling" i.e. freed pointer in this
359: * block, but we use it really as an integer to do some
360: * pointer arithmetic. Insure will raise it as a bug but in
361: * that specific case, that's not !
362: */
363: if (in->base != in->buf->buffer->content) {
364: /*
365: * the buffer has been reallocated
366: */
367: indx = in->cur - in->base;
368: in->base = in->buf->buffer->content;
369: in->cur = &in->buf->buffer->content[indx];
370: }
371: in->end = &in->buf->buffer->content[in->buf->buffer->use];
372:
373: CHECK_BUFFER(in);
374:
375: return(ret);
376: }
377:
378: /**
379: * xmlParserInputShrink:
380: * @in: an XML parser input
381: *
382: * This function removes used input for the parser.
383: */
384: void
385: xmlParserInputShrink(xmlParserInputPtr in) {
386: int used;
387: int ret;
388: int indx;
389:
390: #ifdef DEBUG_INPUT
391: xmlGenericError(xmlGenericErrorContext, "Shrink\n");
392: #endif
393: if (in == NULL) return;
394: if (in->buf == NULL) return;
395: if (in->base == NULL) return;
396: if (in->cur == NULL) return;
397: if (in->buf->buffer == NULL) return;
398:
399: CHECK_BUFFER(in);
400:
401: used = in->cur - in->buf->buffer->content;
402: /*
403: * Do not shrink on large buffers whose only a tiny fraction
404: * was consumed
405: */
406: if (used > INPUT_CHUNK) {
407: ret = xmlBufferShrink(in->buf->buffer, used - LINE_LEN);
408: if (ret > 0) {
409: in->cur -= ret;
410: in->consumed += ret;
411: }
412: in->end = &in->buf->buffer->content[in->buf->buffer->use];
413: }
414:
415: CHECK_BUFFER(in);
416:
417: if (in->buf->buffer->use > INPUT_CHUNK) {
418: return;
419: }
420: xmlParserInputBufferRead(in->buf, 2 * INPUT_CHUNK);
421: if (in->base != in->buf->buffer->content) {
422: /*
423: * the buffer has been reallocated
424: */
425: indx = in->cur - in->base;
426: in->base = in->buf->buffer->content;
427: in->cur = &in->buf->buffer->content[indx];
428: }
429: in->end = &in->buf->buffer->content[in->buf->buffer->use];
430:
431: CHECK_BUFFER(in);
432: }
433:
434: /************************************************************************
435: * *
436: * UTF8 character input and related functions *
437: * *
438: ************************************************************************/
439:
440: /**
441: * xmlNextChar:
442: * @ctxt: the XML parser context
443: *
444: * Skip to the next char input char.
445: */
446:
447: void
448: xmlNextChar(xmlParserCtxtPtr ctxt)
449: {
450: if ((ctxt == NULL) || (ctxt->instate == XML_PARSER_EOF) ||
451: (ctxt->input == NULL))
452: return;
453:
454: if (ctxt->charset == XML_CHAR_ENCODING_UTF8) {
455: if ((*ctxt->input->cur == 0) &&
456: (xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0) &&
457: (ctxt->instate != XML_PARSER_COMMENT)) {
458: /*
459: * If we are at the end of the current entity and
460: * the context allows it, we pop consumed entities
461: * automatically.
462: * the auto closing should be blocked in other cases
463: */
464: xmlPopInput(ctxt);
465: } else {
466: const unsigned char *cur;
467: unsigned char c;
468:
469: /*
470: * 2.11 End-of-Line Handling
471: * the literal two-character sequence "#xD#xA" or a standalone
472: * literal #xD, an XML processor must pass to the application
473: * the single character #xA.
474: */
475: if (*(ctxt->input->cur) == '\n') {
476: ctxt->input->line++; ctxt->input->col = 1;
477: } else
478: ctxt->input->col++;
479:
480: /*
481: * We are supposed to handle UTF8, check it's valid
482: * From rfc2044: encoding of the Unicode values on UTF-8:
483: *
484: * UCS-4 range (hex.) UTF-8 octet sequence (binary)
485: * 0000 0000-0000 007F 0xxxxxxx
486: * 0000 0080-0000 07FF 110xxxxx 10xxxxxx
487: * 0000 0800-0000 FFFF 1110xxxx 10xxxxxx 10xxxxxx
488: *
489: * Check for the 0x110000 limit too
490: */
491: cur = ctxt->input->cur;
492:
493: c = *cur;
494: if (c & 0x80) {
495: if (c == 0xC0)
496: goto encoding_error;
497: if (cur[1] == 0) {
498: xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
499: cur = ctxt->input->cur;
500: }
501: if ((cur[1] & 0xc0) != 0x80)
502: goto encoding_error;
503: if ((c & 0xe0) == 0xe0) {
504: unsigned int val;
505:
506: if (cur[2] == 0) {
507: xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
508: cur = ctxt->input->cur;
509: }
510: if ((cur[2] & 0xc0) != 0x80)
511: goto encoding_error;
512: if ((c & 0xf0) == 0xf0) {
513: if (cur[3] == 0) {
514: xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
515: cur = ctxt->input->cur;
516: }
517: if (((c & 0xf8) != 0xf0) ||
518: ((cur[3] & 0xc0) != 0x80))
519: goto encoding_error;
520: /* 4-byte code */
521: ctxt->input->cur += 4;
522: val = (cur[0] & 0x7) << 18;
523: val |= (cur[1] & 0x3f) << 12;
524: val |= (cur[2] & 0x3f) << 6;
525: val |= cur[3] & 0x3f;
526: } else {
527: /* 3-byte code */
528: ctxt->input->cur += 3;
529: val = (cur[0] & 0xf) << 12;
530: val |= (cur[1] & 0x3f) << 6;
531: val |= cur[2] & 0x3f;
532: }
533: if (((val > 0xd7ff) && (val < 0xe000)) ||
534: ((val > 0xfffd) && (val < 0x10000)) ||
535: (val >= 0x110000)) {
536: xmlErrEncodingInt(ctxt, XML_ERR_INVALID_CHAR,
537: "Char 0x%X out of allowed range\n",
538: val);
539: }
540: } else
541: /* 2-byte code */
542: ctxt->input->cur += 2;
543: } else
544: /* 1-byte code */
545: ctxt->input->cur++;
546:
547: ctxt->nbChars++;
548: if (*ctxt->input->cur == 0)
549: xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
550: }
551: } else {
552: /*
553: * Assume it's a fixed length encoding (1) with
554: * a compatible encoding for the ASCII set, since
555: * XML constructs only use < 128 chars
556: */
557:
558: if (*(ctxt->input->cur) == '\n') {
559: ctxt->input->line++; ctxt->input->col = 1;
560: } else
561: ctxt->input->col++;
562: ctxt->input->cur++;
563: ctxt->nbChars++;
564: if (*ctxt->input->cur == 0)
565: xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
566: }
567: if ((*ctxt->input->cur == '%') && (!ctxt->html))
568: xmlParserHandlePEReference(ctxt);
569: if ((*ctxt->input->cur == 0) &&
570: (xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0))
571: xmlPopInput(ctxt);
572: return;
573: encoding_error:
574: /*
575: * If we detect an UTF8 error that probably mean that the
576: * input encoding didn't get properly advertised in the
577: * declaration header. Report the error and switch the encoding
578: * to ISO-Latin-1 (if you don't like this policy, just declare the
579: * encoding !)
580: */
581: if ((ctxt == NULL) || (ctxt->input == NULL) ||
582: (ctxt->input->end - ctxt->input->cur < 4)) {
583: __xmlErrEncoding(ctxt, XML_ERR_INVALID_CHAR,
584: "Input is not proper UTF-8, indicate encoding !\n",
585: NULL, NULL);
586: } else {
587: char buffer[150];
588:
589: snprintf(buffer, 149, "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n",
590: ctxt->input->cur[0], ctxt->input->cur[1],
591: ctxt->input->cur[2], ctxt->input->cur[3]);
592: __xmlErrEncoding(ctxt, XML_ERR_INVALID_CHAR,
593: "Input is not proper UTF-8, indicate encoding !\n%s",
594: BAD_CAST buffer, NULL);
595: }
596: ctxt->charset = XML_CHAR_ENCODING_8859_1;
597: ctxt->input->cur++;
598: return;
599: }
600:
601: /**
602: * xmlCurrentChar:
603: * @ctxt: the XML parser context
604: * @len: pointer to the length of the char read
605: *
606: * The current char value, if using UTF-8 this may actually span multiple
607: * bytes in the input buffer. Implement the end of line normalization:
608: * 2.11 End-of-Line Handling
609: * Wherever an external parsed entity or the literal entity value
610: * of an internal parsed entity contains either the literal two-character
611: * sequence "#xD#xA" or a standalone literal #xD, an XML processor
612: * must pass to the application the single character #xA.
613: * This behavior can conveniently be produced by normalizing all
614: * line breaks to #xA on input, before parsing.)
615: *
616: * Returns the current char value and its length
617: */
618:
619: int
620: xmlCurrentChar(xmlParserCtxtPtr ctxt, int *len) {
621: if ((ctxt == NULL) || (len == NULL) || (ctxt->input == NULL)) return(0);
622: if (ctxt->instate == XML_PARSER_EOF)
623: return(0);
624:
625: if ((*ctxt->input->cur >= 0x20) && (*ctxt->input->cur <= 0x7F)) {
626: *len = 1;
627: return((int) *ctxt->input->cur);
628: }
629: if (ctxt->charset == XML_CHAR_ENCODING_UTF8) {
630: /*
631: * We are supposed to handle UTF8, check it's valid
632: * From rfc2044: encoding of the Unicode values on UTF-8:
633: *
634: * UCS-4 range (hex.) UTF-8 octet sequence (binary)
635: * 0000 0000-0000 007F 0xxxxxxx
636: * 0000 0080-0000 07FF 110xxxxx 10xxxxxx
637: * 0000 0800-0000 FFFF 1110xxxx 10xxxxxx 10xxxxxx
638: *
639: * Check for the 0x110000 limit too
640: */
641: const unsigned char *cur = ctxt->input->cur;
642: unsigned char c;
643: unsigned int val;
644:
645: c = *cur;
646: if (c & 0x80) {
647: if (((c & 0x40) == 0) || (c == 0xC0))
648: goto encoding_error;
649: if (cur[1] == 0) {
650: xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
651: cur = ctxt->input->cur;
652: }
653: if ((cur[1] & 0xc0) != 0x80)
654: goto encoding_error;
655: if ((c & 0xe0) == 0xe0) {
656: if (cur[2] == 0) {
657: xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
658: cur = ctxt->input->cur;
659: }
660: if ((cur[2] & 0xc0) != 0x80)
661: goto encoding_error;
662: if ((c & 0xf0) == 0xf0) {
663: if (cur[3] == 0) {
664: xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
665: cur = ctxt->input->cur;
666: }
667: if (((c & 0xf8) != 0xf0) ||
668: ((cur[3] & 0xc0) != 0x80))
669: goto encoding_error;
670: /* 4-byte code */
671: *len = 4;
672: val = (cur[0] & 0x7) << 18;
673: val |= (cur[1] & 0x3f) << 12;
674: val |= (cur[2] & 0x3f) << 6;
675: val |= cur[3] & 0x3f;
676: if (val < 0x10000)
677: goto encoding_error;
678: } else {
679: /* 3-byte code */
680: *len = 3;
681: val = (cur[0] & 0xf) << 12;
682: val |= (cur[1] & 0x3f) << 6;
683: val |= cur[2] & 0x3f;
684: if (val < 0x800)
685: goto encoding_error;
686: }
687: } else {
688: /* 2-byte code */
689: *len = 2;
690: val = (cur[0] & 0x1f) << 6;
691: val |= cur[1] & 0x3f;
692: if (val < 0x80)
693: goto encoding_error;
694: }
695: if (!IS_CHAR(val)) {
696: xmlErrEncodingInt(ctxt, XML_ERR_INVALID_CHAR,
697: "Char 0x%X out of allowed range\n", val);
698: }
699: return(val);
700: } else {
701: /* 1-byte code */
702: *len = 1;
703: if (*ctxt->input->cur == 0)
704: xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
705: if ((*ctxt->input->cur == 0) &&
706: (ctxt->input->end > ctxt->input->cur)) {
707: xmlErrEncodingInt(ctxt, XML_ERR_INVALID_CHAR,
708: "Char 0x0 out of allowed range\n", 0);
709: }
710: if (*ctxt->input->cur == 0xD) {
711: if (ctxt->input->cur[1] == 0xA) {
712: ctxt->nbChars++;
713: ctxt->input->cur++;
714: }
715: return(0xA);
716: }
717: return((int) *ctxt->input->cur);
718: }
719: }
720: /*
721: * Assume it's a fixed length encoding (1) with
722: * a compatible encoding for the ASCII set, since
723: * XML constructs only use < 128 chars
724: */
725: *len = 1;
726: if (*ctxt->input->cur == 0xD) {
727: if (ctxt->input->cur[1] == 0xA) {
728: ctxt->nbChars++;
729: ctxt->input->cur++;
730: }
731: return(0xA);
732: }
733: return((int) *ctxt->input->cur);
734: encoding_error:
735: /*
736: * An encoding problem may arise from a truncated input buffer
737: * splitting a character in the middle. In that case do not raise
738: * an error but return 0 to endicate an end of stream problem
739: */
740: if (ctxt->input->end - ctxt->input->cur < 4) {
741: *len = 0;
742: return(0);
743: }
744:
745: /*
746: * If we detect an UTF8 error that probably mean that the
747: * input encoding didn't get properly advertised in the
748: * declaration header. Report the error and switch the encoding
749: * to ISO-Latin-1 (if you don't like this policy, just declare the
750: * encoding !)
751: */
752: {
753: char buffer[150];
754:
755: snprintf(&buffer[0], 149, "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n",
756: ctxt->input->cur[0], ctxt->input->cur[1],
757: ctxt->input->cur[2], ctxt->input->cur[3]);
758: __xmlErrEncoding(ctxt, XML_ERR_INVALID_CHAR,
759: "Input is not proper UTF-8, indicate encoding !\n%s",
760: BAD_CAST buffer, NULL);
761: }
762: ctxt->charset = XML_CHAR_ENCODING_8859_1;
763: *len = 1;
764: return((int) *ctxt->input->cur);
765: }
766:
767: /**
768: * xmlStringCurrentChar:
769: * @ctxt: the XML parser context
770: * @cur: pointer to the beginning of the char
771: * @len: pointer to the length of the char read
772: *
773: * The current char value, if using UTF-8 this may actually span multiple
774: * bytes in the input buffer.
775: *
776: * Returns the current char value and its length
777: */
778:
779: int
780: xmlStringCurrentChar(xmlParserCtxtPtr ctxt, const xmlChar * cur, int *len)
781: {
782: if ((len == NULL) || (cur == NULL)) return(0);
783: if ((ctxt == NULL) || (ctxt->charset == XML_CHAR_ENCODING_UTF8)) {
784: /*
785: * We are supposed to handle UTF8, check it's valid
786: * From rfc2044: encoding of the Unicode values on UTF-8:
787: *
788: * UCS-4 range (hex.) UTF-8 octet sequence (binary)
789: * 0000 0000-0000 007F 0xxxxxxx
790: * 0000 0080-0000 07FF 110xxxxx 10xxxxxx
791: * 0000 0800-0000 FFFF 1110xxxx 10xxxxxx 10xxxxxx
792: *
793: * Check for the 0x110000 limit too
794: */
795: unsigned char c;
796: unsigned int val;
797:
798: c = *cur;
799: if (c & 0x80) {
800: if ((cur[1] & 0xc0) != 0x80)
801: goto encoding_error;
802: if ((c & 0xe0) == 0xe0) {
803:
804: if ((cur[2] & 0xc0) != 0x80)
805: goto encoding_error;
806: if ((c & 0xf0) == 0xf0) {
807: if (((c & 0xf8) != 0xf0) || ((cur[3] & 0xc0) != 0x80))
808: goto encoding_error;
809: /* 4-byte code */
810: *len = 4;
811: val = (cur[0] & 0x7) << 18;
812: val |= (cur[1] & 0x3f) << 12;
813: val |= (cur[2] & 0x3f) << 6;
814: val |= cur[3] & 0x3f;
815: } else {
816: /* 3-byte code */
817: *len = 3;
818: val = (cur[0] & 0xf) << 12;
819: val |= (cur[1] & 0x3f) << 6;
820: val |= cur[2] & 0x3f;
821: }
822: } else {
823: /* 2-byte code */
824: *len = 2;
825: val = (cur[0] & 0x1f) << 6;
826: val |= cur[1] & 0x3f;
827: }
828: if (!IS_CHAR(val)) {
829: xmlErrEncodingInt(ctxt, XML_ERR_INVALID_CHAR,
830: "Char 0x%X out of allowed range\n", val);
831: }
832: return (val);
833: } else {
834: /* 1-byte code */
835: *len = 1;
836: return ((int) *cur);
837: }
838: }
839: /*
840: * Assume it's a fixed length encoding (1) with
841: * a compatible encoding for the ASCII set, since
842: * XML constructs only use < 128 chars
843: */
844: *len = 1;
845: return ((int) *cur);
846: encoding_error:
847:
848: /*
849: * An encoding problem may arise from a truncated input buffer
850: * splitting a character in the middle. In that case do not raise
851: * an error but return 0 to endicate an end of stream problem
852: */
853: if ((ctxt == NULL) || (ctxt->input == NULL) ||
854: (ctxt->input->end - ctxt->input->cur < 4)) {
855: *len = 0;
856: return(0);
857: }
858: /*
859: * If we detect an UTF8 error that probably mean that the
860: * input encoding didn't get properly advertised in the
861: * declaration header. Report the error and switch the encoding
862: * to ISO-Latin-1 (if you don't like this policy, just declare the
863: * encoding !)
864: */
865: {
866: char buffer[150];
867:
868: snprintf(buffer, 149, "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n",
869: ctxt->input->cur[0], ctxt->input->cur[1],
870: ctxt->input->cur[2], ctxt->input->cur[3]);
871: __xmlErrEncoding(ctxt, XML_ERR_INVALID_CHAR,
872: "Input is not proper UTF-8, indicate encoding !\n%s",
873: BAD_CAST buffer, NULL);
874: }
875: *len = 1;
876: return ((int) *cur);
877: }
878:
879: /**
880: * xmlCopyCharMultiByte:
881: * @out: pointer to an array of xmlChar
882: * @val: the char value
883: *
884: * append the char value in the array
885: *
886: * Returns the number of xmlChar written
887: */
888: int
889: xmlCopyCharMultiByte(xmlChar *out, int val) {
890: if (out == NULL) return(0);
891: /*
892: * We are supposed to handle UTF8, check it's valid
893: * From rfc2044: encoding of the Unicode values on UTF-8:
894: *
895: * UCS-4 range (hex.) UTF-8 octet sequence (binary)
896: * 0000 0000-0000 007F 0xxxxxxx
897: * 0000 0080-0000 07FF 110xxxxx 10xxxxxx
898: * 0000 0800-0000 FFFF 1110xxxx 10xxxxxx 10xxxxxx
899: */
900: if (val >= 0x80) {
901: xmlChar *savedout = out;
902: int bits;
903: if (val < 0x800) { *out++= (val >> 6) | 0xC0; bits= 0; }
904: else if (val < 0x10000) { *out++= (val >> 12) | 0xE0; bits= 6;}
905: else if (val < 0x110000) { *out++= (val >> 18) | 0xF0; bits= 12; }
906: else {
907: xmlErrEncodingInt(NULL, XML_ERR_INVALID_CHAR,
908: "Internal error, xmlCopyCharMultiByte 0x%X out of bound\n",
909: val);
910: return(0);
911: }
912: for ( ; bits >= 0; bits-= 6)
913: *out++= ((val >> bits) & 0x3F) | 0x80 ;
914: return (out - savedout);
915: }
916: *out = (xmlChar) val;
917: return 1;
918: }
919:
920: /**
921: * xmlCopyChar:
922: * @len: Ignored, compatibility
923: * @out: pointer to an array of xmlChar
924: * @val: the char value
925: *
926: * append the char value in the array
927: *
928: * Returns the number of xmlChar written
929: */
930:
931: int
932: xmlCopyChar(int len ATTRIBUTE_UNUSED, xmlChar *out, int val) {
933: if (out == NULL) return(0);
934: /* the len parameter is ignored */
935: if (val >= 0x80) {
936: return(xmlCopyCharMultiByte (out, val));
937: }
938: *out = (xmlChar) val;
939: return 1;
940: }
941:
942: /************************************************************************
943: * *
944: * Commodity functions to switch encodings *
945: * *
946: ************************************************************************/
947:
948: /* defined in encoding.c, not public */
949: int
950: xmlCharEncFirstLineInt(xmlCharEncodingHandler *handler, xmlBufferPtr out,
951: xmlBufferPtr in, int len);
952:
953: static int
954: xmlSwitchToEncodingInt(xmlParserCtxtPtr ctxt,
955: xmlCharEncodingHandlerPtr handler, int len);
956: static int
957: xmlSwitchInputEncodingInt(xmlParserCtxtPtr ctxt, xmlParserInputPtr input,
958: xmlCharEncodingHandlerPtr handler, int len);
959: /**
960: * xmlSwitchEncoding:
961: * @ctxt: the parser context
962: * @enc: the encoding value (number)
963: *
964: * change the input functions when discovering the character encoding
965: * of a given entity.
966: *
967: * Returns 0 in case of success, -1 otherwise
968: */
969: int
970: xmlSwitchEncoding(xmlParserCtxtPtr ctxt, xmlCharEncoding enc)
971: {
972: xmlCharEncodingHandlerPtr handler;
973: int len = -1;
974:
975: if (ctxt == NULL) return(-1);
976: switch (enc) {
977: case XML_CHAR_ENCODING_ERROR:
978: __xmlErrEncoding(ctxt, XML_ERR_UNKNOWN_ENCODING,
979: "encoding unknown\n", NULL, NULL);
980: return(-1);
981: case XML_CHAR_ENCODING_NONE:
982: /* let's assume it's UTF-8 without the XML decl */
983: ctxt->charset = XML_CHAR_ENCODING_UTF8;
984: return(0);
985: case XML_CHAR_ENCODING_UTF8:
986: /* default encoding, no conversion should be needed */
987: ctxt->charset = XML_CHAR_ENCODING_UTF8;
988:
989: /*
990: * Errata on XML-1.0 June 20 2001
991: * Specific handling of the Byte Order Mark for
992: * UTF-8
993: */
994: if ((ctxt->input != NULL) &&
995: (ctxt->input->cur[0] == 0xEF) &&
996: (ctxt->input->cur[1] == 0xBB) &&
997: (ctxt->input->cur[2] == 0xBF)) {
998: ctxt->input->cur += 3;
999: }
1000: return(0);
1001: case XML_CHAR_ENCODING_UTF16LE:
1002: case XML_CHAR_ENCODING_UTF16BE:
1003: /*The raw input characters are encoded
1004: *in UTF-16. As we expect this function
1005: *to be called after xmlCharEncInFunc, we expect
1006: *ctxt->input->cur to contain UTF-8 encoded characters.
1007: *So the raw UTF16 Byte Order Mark
1008: *has also been converted into
1009: *an UTF-8 BOM. Let's skip that BOM.
1010: */
1011: if ((ctxt->input != NULL) && (ctxt->input->cur != NULL) &&
1012: (ctxt->input->cur[0] == 0xEF) &&
1013: (ctxt->input->cur[1] == 0xBB) &&
1014: (ctxt->input->cur[2] == 0xBF)) {
1015: ctxt->input->cur += 3;
1016: }
1017: len = 90;
1018: break;
1019: case XML_CHAR_ENCODING_UCS2:
1020: len = 90;
1021: break;
1022: case XML_CHAR_ENCODING_UCS4BE:
1023: case XML_CHAR_ENCODING_UCS4LE:
1024: case XML_CHAR_ENCODING_UCS4_2143:
1025: case XML_CHAR_ENCODING_UCS4_3412:
1026: len = 180;
1027: break;
1028: case XML_CHAR_ENCODING_EBCDIC:
1029: case XML_CHAR_ENCODING_8859_1:
1030: case XML_CHAR_ENCODING_8859_2:
1031: case XML_CHAR_ENCODING_8859_3:
1032: case XML_CHAR_ENCODING_8859_4:
1033: case XML_CHAR_ENCODING_8859_5:
1034: case XML_CHAR_ENCODING_8859_6:
1035: case XML_CHAR_ENCODING_8859_7:
1036: case XML_CHAR_ENCODING_8859_8:
1037: case XML_CHAR_ENCODING_8859_9:
1038: case XML_CHAR_ENCODING_ASCII:
1039: case XML_CHAR_ENCODING_2022_JP:
1040: case XML_CHAR_ENCODING_SHIFT_JIS:
1041: case XML_CHAR_ENCODING_EUC_JP:
1042: len = 45;
1043: break;
1044: }
1045: handler = xmlGetCharEncodingHandler(enc);
1046: if (handler == NULL) {
1047: /*
1048: * Default handlers.
1049: */
1050: switch (enc) {
1051: case XML_CHAR_ENCODING_ASCII:
1052: /* default encoding, no conversion should be needed */
1053: ctxt->charset = XML_CHAR_ENCODING_UTF8;
1054: return(0);
1055: case XML_CHAR_ENCODING_UTF16LE:
1056: break;
1057: case XML_CHAR_ENCODING_UTF16BE:
1058: break;
1059: case XML_CHAR_ENCODING_UCS4LE:
1060: __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
1061: "encoding not supported %s\n",
1062: BAD_CAST "USC4 little endian", NULL);
1063: break;
1064: case XML_CHAR_ENCODING_UCS4BE:
1065: __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
1066: "encoding not supported %s\n",
1067: BAD_CAST "USC4 big endian", NULL);
1068: break;
1069: case XML_CHAR_ENCODING_EBCDIC:
1070: __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
1071: "encoding not supported %s\n",
1072: BAD_CAST "EBCDIC", NULL);
1073: break;
1074: case XML_CHAR_ENCODING_UCS4_2143:
1075: __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
1076: "encoding not supported %s\n",
1077: BAD_CAST "UCS4 2143", NULL);
1078: break;
1079: case XML_CHAR_ENCODING_UCS4_3412:
1080: __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
1081: "encoding not supported %s\n",
1082: BAD_CAST "UCS4 3412", NULL);
1083: break;
1084: case XML_CHAR_ENCODING_UCS2:
1085: __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
1086: "encoding not supported %s\n",
1087: BAD_CAST "UCS2", NULL);
1088: break;
1089: case XML_CHAR_ENCODING_8859_1:
1090: case XML_CHAR_ENCODING_8859_2:
1091: case XML_CHAR_ENCODING_8859_3:
1092: case XML_CHAR_ENCODING_8859_4:
1093: case XML_CHAR_ENCODING_8859_5:
1094: case XML_CHAR_ENCODING_8859_6:
1095: case XML_CHAR_ENCODING_8859_7:
1096: case XML_CHAR_ENCODING_8859_8:
1097: case XML_CHAR_ENCODING_8859_9:
1098: /*
1099: * We used to keep the internal content in the
1100: * document encoding however this turns being unmaintainable
1101: * So xmlGetCharEncodingHandler() will return non-null
1102: * values for this now.
1103: */
1104: if ((ctxt->inputNr == 1) &&
1105: (ctxt->encoding == NULL) &&
1106: (ctxt->input != NULL) &&
1107: (ctxt->input->encoding != NULL)) {
1108: ctxt->encoding = xmlStrdup(ctxt->input->encoding);
1109: }
1110: ctxt->charset = enc;
1111: return(0);
1112: case XML_CHAR_ENCODING_2022_JP:
1113: __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
1114: "encoding not supported %s\n",
1115: BAD_CAST "ISO-2022-JP", NULL);
1116: break;
1117: case XML_CHAR_ENCODING_SHIFT_JIS:
1118: __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
1119: "encoding not supported %s\n",
1120: BAD_CAST "Shift_JIS", NULL);
1121: break;
1122: case XML_CHAR_ENCODING_EUC_JP:
1123: __xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
1124: "encoding not supported %s\n",
1125: BAD_CAST "EUC-JP", NULL);
1126: break;
1127: default:
1128: break;
1129: }
1130: }
1131: if (handler == NULL)
1132: return(-1);
1133: ctxt->charset = XML_CHAR_ENCODING_UTF8;
1134: return(xmlSwitchToEncodingInt(ctxt, handler, len));
1135: }
1136:
1137: /**
1138: * xmlSwitchInputEncoding:
1139: * @ctxt: the parser context
1140: * @input: the input stream
1141: * @handler: the encoding handler
1142: * @len: the number of bytes to convert for the first line or -1
1143: *
1144: * change the input functions when discovering the character encoding
1145: * of a given entity.
1146: *
1147: * Returns 0 in case of success, -1 otherwise
1148: */
1149: static int
1150: xmlSwitchInputEncodingInt(xmlParserCtxtPtr ctxt, xmlParserInputPtr input,
1151: xmlCharEncodingHandlerPtr handler, int len)
1152: {
1153: int nbchars;
1154:
1155: if (handler == NULL)
1156: return (-1);
1157: if (input == NULL)
1158: return (-1);
1159: if (input->buf != NULL) {
1160: if (input->buf->encoder != NULL) {
1161: /*
1162: * Check in case the auto encoding detetection triggered
1163: * in already.
1164: */
1165: if (input->buf->encoder == handler)
1166: return (0);
1167:
1168: /*
1169: * "UTF-16" can be used for both LE and BE
1170: if ((!xmlStrncmp(BAD_CAST input->buf->encoder->name,
1171: BAD_CAST "UTF-16", 6)) &&
1172: (!xmlStrncmp(BAD_CAST handler->name,
1173: BAD_CAST "UTF-16", 6))) {
1174: return(0);
1175: }
1176: */
1177:
1178: /*
1179: * Note: this is a bit dangerous, but that's what it
1180: * takes to use nearly compatible signature for different
1181: * encodings.
1182: */
1183: xmlCharEncCloseFunc(input->buf->encoder);
1184: input->buf->encoder = handler;
1185: return (0);
1186: }
1187: input->buf->encoder = handler;
1188:
1189: /*
1190: * Is there already some content down the pipe to convert ?
1191: */
1192: if ((input->buf->buffer != NULL) && (input->buf->buffer->use > 0)) {
1193: int processed;
1194: unsigned int use;
1195:
1196: /*
1197: * Specific handling of the Byte Order Mark for
1198: * UTF-16
1199: */
1200: if ((handler->name != NULL) &&
1201: (!strcmp(handler->name, "UTF-16LE") ||
1202: !strcmp(handler->name, "UTF-16")) &&
1203: (input->cur[0] == 0xFF) && (input->cur[1] == 0xFE)) {
1204: input->cur += 2;
1205: }
1206: if ((handler->name != NULL) &&
1207: (!strcmp(handler->name, "UTF-16BE")) &&
1208: (input->cur[0] == 0xFE) && (input->cur[1] == 0xFF)) {
1209: input->cur += 2;
1210: }
1211: /*
1212: * Errata on XML-1.0 June 20 2001
1213: * Specific handling of the Byte Order Mark for
1214: * UTF-8
1215: */
1216: if ((handler->name != NULL) &&
1217: (!strcmp(handler->name, "UTF-8")) &&
1218: (input->cur[0] == 0xEF) &&
1219: (input->cur[1] == 0xBB) && (input->cur[2] == 0xBF)) {
1220: input->cur += 3;
1221: }
1222:
1223: /*
1224: * Shrink the current input buffer.
1225: * Move it as the raw buffer and create a new input buffer
1226: */
1227: processed = input->cur - input->base;
1228: xmlBufferShrink(input->buf->buffer, processed);
1229: input->buf->raw = input->buf->buffer;
1230: input->buf->buffer = xmlBufferCreate();
1231: input->buf->rawconsumed = processed;
1232: use = input->buf->raw->use;
1233:
1234: if (ctxt->html) {
1235: /*
1236: * convert as much as possible of the buffer
1237: */
1238: nbchars = xmlCharEncInFunc(input->buf->encoder,
1239: input->buf->buffer,
1240: input->buf->raw);
1241: } else {
1242: /*
1243: * convert just enough to get
1244: * '<?xml version="1.0" encoding="xxx"?>'
1245: * parsed with the autodetected encoding
1246: * into the parser reading buffer.
1247: */
1248: nbchars = xmlCharEncFirstLineInt(input->buf->encoder,
1249: input->buf->buffer,
1250: input->buf->raw,
1251: len);
1252: }
1253: if (nbchars < 0) {
1254: xmlErrInternal(ctxt,
1255: "switching encoding: encoder error\n",
1256: NULL);
1257: return (-1);
1258: }
1259: input->buf->rawconsumed += use - input->buf->raw->use;
1260: input->base = input->cur = input->buf->buffer->content;
1261: input->end = &input->base[input->buf->buffer->use];
1262:
1263: }
1264: return (0);
1265: } else if (input->length == 0) {
1266: /*
1267: * When parsing a static memory array one must know the
1268: * size to be able to convert the buffer.
1269: */
1270: xmlErrInternal(ctxt, "switching encoding : no input\n", NULL);
1271: return (-1);
1272: }
1273: return (0);
1274: }
1275:
1276: /**
1277: * xmlSwitchInputEncoding:
1278: * @ctxt: the parser context
1279: * @input: the input stream
1280: * @handler: the encoding handler
1281: *
1282: * change the input functions when discovering the character encoding
1283: * of a given entity.
1284: *
1285: * Returns 0 in case of success, -1 otherwise
1286: */
1287: int
1288: xmlSwitchInputEncoding(xmlParserCtxtPtr ctxt, xmlParserInputPtr input,
1289: xmlCharEncodingHandlerPtr handler) {
1290: return(xmlSwitchInputEncodingInt(ctxt, input, handler, -1));
1291: }
1292:
1293: /**
1294: * xmlSwitchToEncodingInt:
1295: * @ctxt: the parser context
1296: * @handler: the encoding handler
1297: * @len: the lenght to convert or -1
1298: *
1299: * change the input functions when discovering the character encoding
1300: * of a given entity, and convert only @len bytes of the output, this
1301: * is needed on auto detect to allows any declared encoding later to
1302: * convert the actual content after the xmlDecl
1303: *
1304: * Returns 0 in case of success, -1 otherwise
1305: */
1306: static int
1307: xmlSwitchToEncodingInt(xmlParserCtxtPtr ctxt,
1308: xmlCharEncodingHandlerPtr handler, int len) {
1309: int ret = 0;
1310:
1311: if (handler != NULL) {
1312: if (ctxt->input != NULL) {
1313: ret = xmlSwitchInputEncodingInt(ctxt, ctxt->input, handler, len);
1314: } else {
1315: xmlErrInternal(ctxt, "xmlSwitchToEncoding : no input\n",
1316: NULL);
1317: return(-1);
1318: }
1319: /*
1320: * The parsing is now done in UTF8 natively
1321: */
1322: ctxt->charset = XML_CHAR_ENCODING_UTF8;
1323: } else
1324: return(-1);
1325: return(ret);
1326: }
1327:
1328: /**
1329: * xmlSwitchToEncoding:
1330: * @ctxt: the parser context
1331: * @handler: the encoding handler
1332: *
1333: * change the input functions when discovering the character encoding
1334: * of a given entity.
1335: *
1336: * Returns 0 in case of success, -1 otherwise
1337: */
1338: int
1339: xmlSwitchToEncoding(xmlParserCtxtPtr ctxt, xmlCharEncodingHandlerPtr handler)
1340: {
1341: return (xmlSwitchToEncodingInt(ctxt, handler, -1));
1342: }
1343:
1344: /************************************************************************
1345: * *
1346: * Commodity functions to handle entities processing *
1347: * *
1348: ************************************************************************/
1349:
1350: /**
1351: * xmlFreeInputStream:
1352: * @input: an xmlParserInputPtr
1353: *
1354: * Free up an input stream.
1355: */
1356: void
1357: xmlFreeInputStream(xmlParserInputPtr input) {
1358: if (input == NULL) return;
1359:
1360: if (input->filename != NULL) xmlFree((char *) input->filename);
1361: if (input->directory != NULL) xmlFree((char *) input->directory);
1362: if (input->encoding != NULL) xmlFree((char *) input->encoding);
1363: if (input->version != NULL) xmlFree((char *) input->version);
1364: if ((input->free != NULL) && (input->base != NULL))
1365: input->free((xmlChar *) input->base);
1366: if (input->buf != NULL)
1367: xmlFreeParserInputBuffer(input->buf);
1368: xmlFree(input);
1369: }
1370:
1371: /**
1372: * xmlNewInputStream:
1373: * @ctxt: an XML parser context
1374: *
1.1.1.2 ! misho 1375: * Create a new input stream structure.
! 1376: *
1.1 misho 1377: * Returns the new input stream or NULL
1378: */
1379: xmlParserInputPtr
1380: xmlNewInputStream(xmlParserCtxtPtr ctxt) {
1381: xmlParserInputPtr input;
1382:
1383: input = (xmlParserInputPtr) xmlMalloc(sizeof(xmlParserInput));
1384: if (input == NULL) {
1385: xmlErrMemory(ctxt, "couldn't allocate a new input stream\n");
1386: return(NULL);
1387: }
1388: memset(input, 0, sizeof(xmlParserInput));
1389: input->line = 1;
1390: input->col = 1;
1391: input->standalone = -1;
1.1.1.2 ! misho 1392:
1.1 misho 1393: /*
1.1.1.2 ! misho 1394: * If the context is NULL the id cannot be initialized, but that
! 1395: * should not happen while parsing which is the situation where
! 1396: * the id is actually needed.
1.1 misho 1397: */
1.1.1.2 ! misho 1398: if (ctxt != NULL)
! 1399: input->id = ctxt->input_id++;
! 1400:
1.1 misho 1401: return(input);
1402: }
1403:
1404: /**
1405: * xmlNewIOInputStream:
1406: * @ctxt: an XML parser context
1407: * @input: an I/O Input
1408: * @enc: the charset encoding if known
1409: *
1410: * Create a new input stream structure encapsulating the @input into
1411: * a stream suitable for the parser.
1412: *
1413: * Returns the new input stream or NULL
1414: */
1415: xmlParserInputPtr
1416: xmlNewIOInputStream(xmlParserCtxtPtr ctxt, xmlParserInputBufferPtr input,
1417: xmlCharEncoding enc) {
1418: xmlParserInputPtr inputStream;
1419:
1420: if (input == NULL) return(NULL);
1421: if (xmlParserDebugEntities)
1422: xmlGenericError(xmlGenericErrorContext, "new input from I/O\n");
1423: inputStream = xmlNewInputStream(ctxt);
1424: if (inputStream == NULL) {
1425: return(NULL);
1426: }
1427: inputStream->filename = NULL;
1428: inputStream->buf = input;
1429: inputStream->base = inputStream->buf->buffer->content;
1430: inputStream->cur = inputStream->buf->buffer->content;
1431: inputStream->end = &inputStream->base[inputStream->buf->buffer->use];
1432: if (enc != XML_CHAR_ENCODING_NONE) {
1433: xmlSwitchEncoding(ctxt, enc);
1434: }
1435:
1436: return(inputStream);
1437: }
1438:
1439: /**
1440: * xmlNewEntityInputStream:
1441: * @ctxt: an XML parser context
1442: * @entity: an Entity pointer
1443: *
1444: * Create a new input stream based on an xmlEntityPtr
1445: *
1446: * Returns the new input stream or NULL
1447: */
1448: xmlParserInputPtr
1449: xmlNewEntityInputStream(xmlParserCtxtPtr ctxt, xmlEntityPtr entity) {
1450: xmlParserInputPtr input;
1451:
1452: if (entity == NULL) {
1453: xmlErrInternal(ctxt, "xmlNewEntityInputStream entity = NULL\n",
1454: NULL);
1455: return(NULL);
1456: }
1457: if (xmlParserDebugEntities)
1458: xmlGenericError(xmlGenericErrorContext,
1459: "new input from entity: %s\n", entity->name);
1460: if (entity->content == NULL) {
1461: switch (entity->etype) {
1462: case XML_EXTERNAL_GENERAL_UNPARSED_ENTITY:
1463: xmlErrInternal(ctxt, "Cannot parse entity %s\n",
1464: entity->name);
1465: break;
1466: case XML_EXTERNAL_GENERAL_PARSED_ENTITY:
1467: case XML_EXTERNAL_PARAMETER_ENTITY:
1468: return(xmlLoadExternalEntity((char *) entity->URI,
1469: (char *) entity->ExternalID, ctxt));
1470: case XML_INTERNAL_GENERAL_ENTITY:
1471: xmlErrInternal(ctxt,
1472: "Internal entity %s without content !\n",
1473: entity->name);
1474: break;
1475: case XML_INTERNAL_PARAMETER_ENTITY:
1476: xmlErrInternal(ctxt,
1477: "Internal parameter entity %s without content !\n",
1478: entity->name);
1479: break;
1480: case XML_INTERNAL_PREDEFINED_ENTITY:
1481: xmlErrInternal(ctxt,
1482: "Predefined entity %s without content !\n",
1483: entity->name);
1484: break;
1485: }
1486: return(NULL);
1487: }
1488: input = xmlNewInputStream(ctxt);
1489: if (input == NULL) {
1490: return(NULL);
1491: }
1492: if (entity->URI != NULL)
1493: input->filename = (char *) xmlStrdup((xmlChar *) entity->URI);
1494: input->base = entity->content;
1495: input->cur = entity->content;
1496: input->length = entity->length;
1497: input->end = &entity->content[input->length];
1498: return(input);
1499: }
1500:
1501: /**
1502: * xmlNewStringInputStream:
1503: * @ctxt: an XML parser context
1504: * @buffer: an memory buffer
1505: *
1506: * Create a new input stream based on a memory buffer.
1507: * Returns the new input stream
1508: */
1509: xmlParserInputPtr
1510: xmlNewStringInputStream(xmlParserCtxtPtr ctxt, const xmlChar *buffer) {
1511: xmlParserInputPtr input;
1512:
1513: if (buffer == NULL) {
1514: xmlErrInternal(ctxt, "xmlNewStringInputStream string = NULL\n",
1515: NULL);
1516: return(NULL);
1517: }
1518: if (xmlParserDebugEntities)
1519: xmlGenericError(xmlGenericErrorContext,
1520: "new fixed input: %.30s\n", buffer);
1521: input = xmlNewInputStream(ctxt);
1522: if (input == NULL) {
1523: xmlErrMemory(ctxt, "couldn't allocate a new input stream\n");
1524: return(NULL);
1525: }
1526: input->base = buffer;
1527: input->cur = buffer;
1528: input->length = xmlStrlen(buffer);
1529: input->end = &buffer[input->length];
1530: return(input);
1531: }
1532:
1533: /**
1534: * xmlNewInputFromFile:
1535: * @ctxt: an XML parser context
1536: * @filename: the filename to use as entity
1537: *
1538: * Create a new input stream based on a file or an URL.
1539: *
1540: * Returns the new input stream or NULL in case of error
1541: */
1542: xmlParserInputPtr
1543: xmlNewInputFromFile(xmlParserCtxtPtr ctxt, const char *filename) {
1544: xmlParserInputBufferPtr buf;
1545: xmlParserInputPtr inputStream;
1546: char *directory = NULL;
1547: xmlChar *URI = NULL;
1548:
1549: if (xmlParserDebugEntities)
1550: xmlGenericError(xmlGenericErrorContext,
1551: "new input from file: %s\n", filename);
1552: if (ctxt == NULL) return(NULL);
1553: buf = xmlParserInputBufferCreateFilename(filename, XML_CHAR_ENCODING_NONE);
1554: if (buf == NULL) {
1555: if (filename == NULL)
1556: __xmlLoaderErr(ctxt,
1557: "failed to load external entity: NULL filename \n",
1558: NULL);
1559: else
1560: __xmlLoaderErr(ctxt, "failed to load external entity \"%s\"\n",
1561: (const char *) filename);
1562: return(NULL);
1563: }
1564:
1565: inputStream = xmlNewInputStream(ctxt);
1566: if (inputStream == NULL)
1567: return(NULL);
1568:
1569: inputStream->buf = buf;
1570: inputStream = xmlCheckHTTPInput(ctxt, inputStream);
1571: if (inputStream == NULL)
1572: return(NULL);
1573:
1574: if (inputStream->filename == NULL)
1575: URI = xmlStrdup((xmlChar *) filename);
1576: else
1577: URI = xmlStrdup((xmlChar *) inputStream->filename);
1578: directory = xmlParserGetDirectory((const char *) URI);
1579: if (inputStream->filename != NULL) xmlFree((char *)inputStream->filename);
1580: inputStream->filename = (char *) xmlCanonicPath((const xmlChar *) URI);
1581: if (URI != NULL) xmlFree((char *) URI);
1582: inputStream->directory = directory;
1583:
1584: inputStream->base = inputStream->buf->buffer->content;
1585: inputStream->cur = inputStream->buf->buffer->content;
1586: inputStream->end = &inputStream->base[inputStream->buf->buffer->use];
1587: if ((ctxt->directory == NULL) && (directory != NULL))
1588: ctxt->directory = (char *) xmlStrdup((const xmlChar *) directory);
1589: return(inputStream);
1590: }
1591:
1592: /************************************************************************
1593: * *
1594: * Commodity functions to handle parser contexts *
1595: * *
1596: ************************************************************************/
1597:
1598: /**
1599: * xmlInitParserCtxt:
1600: * @ctxt: an XML parser context
1601: *
1602: * Initialize a parser context
1603: *
1604: * Returns 0 in case of success and -1 in case of error
1605: */
1606:
1607: int
1608: xmlInitParserCtxt(xmlParserCtxtPtr ctxt)
1609: {
1610: xmlParserInputPtr input;
1611:
1612: if(ctxt==NULL) {
1613: xmlErrInternal(NULL, "Got NULL parser context\n", NULL);
1614: return(-1);
1615: }
1616:
1617: xmlDefaultSAXHandlerInit();
1618:
1619: if (ctxt->dict == NULL)
1620: ctxt->dict = xmlDictCreate();
1621: if (ctxt->dict == NULL) {
1622: xmlErrMemory(NULL, "cannot initialize parser context\n");
1623: return(-1);
1624: }
1625: if (ctxt->sax == NULL)
1626: ctxt->sax = (xmlSAXHandler *) xmlMalloc(sizeof(xmlSAXHandler));
1627: if (ctxt->sax == NULL) {
1628: xmlErrMemory(NULL, "cannot initialize parser context\n");
1629: return(-1);
1630: }
1631: else
1632: xmlSAXVersion(ctxt->sax, 2);
1633:
1634: ctxt->maxatts = 0;
1635: ctxt->atts = NULL;
1636: /* Allocate the Input stack */
1637: if (ctxt->inputTab == NULL) {
1638: ctxt->inputTab = (xmlParserInputPtr *)
1639: xmlMalloc(5 * sizeof(xmlParserInputPtr));
1640: ctxt->inputMax = 5;
1641: }
1642: if (ctxt->inputTab == NULL) {
1643: xmlErrMemory(NULL, "cannot initialize parser context\n");
1644: ctxt->inputNr = 0;
1645: ctxt->inputMax = 0;
1646: ctxt->input = NULL;
1647: return(-1);
1648: }
1649: while ((input = inputPop(ctxt)) != NULL) { /* Non consuming */
1650: xmlFreeInputStream(input);
1651: }
1652: ctxt->inputNr = 0;
1653: ctxt->input = NULL;
1654:
1655: ctxt->version = NULL;
1656: ctxt->encoding = NULL;
1657: ctxt->standalone = -1;
1658: ctxt->hasExternalSubset = 0;
1659: ctxt->hasPErefs = 0;
1660: ctxt->html = 0;
1661: ctxt->external = 0;
1662: ctxt->instate = XML_PARSER_START;
1663: ctxt->token = 0;
1664: ctxt->directory = NULL;
1665:
1666: /* Allocate the Node stack */
1667: if (ctxt->nodeTab == NULL) {
1668: ctxt->nodeTab = (xmlNodePtr *) xmlMalloc(10 * sizeof(xmlNodePtr));
1669: ctxt->nodeMax = 10;
1670: }
1671: if (ctxt->nodeTab == NULL) {
1672: xmlErrMemory(NULL, "cannot initialize parser context\n");
1673: ctxt->nodeNr = 0;
1674: ctxt->nodeMax = 0;
1675: ctxt->node = NULL;
1676: ctxt->inputNr = 0;
1677: ctxt->inputMax = 0;
1678: ctxt->input = NULL;
1679: return(-1);
1680: }
1681: ctxt->nodeNr = 0;
1682: ctxt->node = NULL;
1683:
1684: /* Allocate the Name stack */
1685: if (ctxt->nameTab == NULL) {
1686: ctxt->nameTab = (const xmlChar **) xmlMalloc(10 * sizeof(xmlChar *));
1687: ctxt->nameMax = 10;
1688: }
1689: if (ctxt->nameTab == NULL) {
1690: xmlErrMemory(NULL, "cannot initialize parser context\n");
1691: ctxt->nodeNr = 0;
1692: ctxt->nodeMax = 0;
1693: ctxt->node = NULL;
1694: ctxt->inputNr = 0;
1695: ctxt->inputMax = 0;
1696: ctxt->input = NULL;
1697: ctxt->nameNr = 0;
1698: ctxt->nameMax = 0;
1699: ctxt->name = NULL;
1700: return(-1);
1701: }
1702: ctxt->nameNr = 0;
1703: ctxt->name = NULL;
1704:
1705: /* Allocate the space stack */
1706: if (ctxt->spaceTab == NULL) {
1707: ctxt->spaceTab = (int *) xmlMalloc(10 * sizeof(int));
1708: ctxt->spaceMax = 10;
1709: }
1710: if (ctxt->spaceTab == NULL) {
1711: xmlErrMemory(NULL, "cannot initialize parser context\n");
1712: ctxt->nodeNr = 0;
1713: ctxt->nodeMax = 0;
1714: ctxt->node = NULL;
1715: ctxt->inputNr = 0;
1716: ctxt->inputMax = 0;
1717: ctxt->input = NULL;
1718: ctxt->nameNr = 0;
1719: ctxt->nameMax = 0;
1720: ctxt->name = NULL;
1721: ctxt->spaceNr = 0;
1722: ctxt->spaceMax = 0;
1723: ctxt->space = NULL;
1724: return(-1);
1725: }
1726: ctxt->spaceNr = 1;
1727: ctxt->spaceMax = 10;
1728: ctxt->spaceTab[0] = -1;
1729: ctxt->space = &ctxt->spaceTab[0];
1730: ctxt->userData = ctxt;
1731: ctxt->myDoc = NULL;
1732: ctxt->wellFormed = 1;
1733: ctxt->nsWellFormed = 1;
1734: ctxt->valid = 1;
1735: ctxt->loadsubset = xmlLoadExtDtdDefaultValue;
1736: ctxt->validate = xmlDoValidityCheckingDefaultValue;
1737: ctxt->pedantic = xmlPedanticParserDefaultValue;
1738: ctxt->linenumbers = xmlLineNumbersDefaultValue;
1739: ctxt->keepBlanks = xmlKeepBlanksDefaultValue;
1740: if (ctxt->keepBlanks == 0)
1741: ctxt->sax->ignorableWhitespace = xmlSAX2IgnorableWhitespace;
1742:
1743: ctxt->vctxt.finishDtd = XML_CTXT_FINISH_DTD_0;
1744: ctxt->vctxt.userData = ctxt;
1745: ctxt->vctxt.error = xmlParserValidityError;
1746: ctxt->vctxt.warning = xmlParserValidityWarning;
1747: if (ctxt->validate) {
1748: if (xmlGetWarningsDefaultValue == 0)
1749: ctxt->vctxt.warning = NULL;
1750: else
1751: ctxt->vctxt.warning = xmlParserValidityWarning;
1752: ctxt->vctxt.nodeMax = 0;
1753: }
1754: ctxt->replaceEntities = xmlSubstituteEntitiesDefaultValue;
1755: ctxt->record_info = 0;
1756: ctxt->nbChars = 0;
1757: ctxt->checkIndex = 0;
1758: ctxt->inSubset = 0;
1759: ctxt->errNo = XML_ERR_OK;
1760: ctxt->depth = 0;
1761: ctxt->charset = XML_CHAR_ENCODING_UTF8;
1762: ctxt->catalogs = NULL;
1763: ctxt->nbentities = 0;
1.1.1.2 ! misho 1764: ctxt->input_id = 1;
1.1 misho 1765: xmlInitNodeInfoSeq(&ctxt->node_seq);
1766: return(0);
1767: }
1768:
1769: /**
1770: * xmlFreeParserCtxt:
1771: * @ctxt: an XML parser context
1772: *
1773: * Free all the memory used by a parser context. However the parsed
1774: * document in ctxt->myDoc is not freed.
1775: */
1776:
1777: void
1778: xmlFreeParserCtxt(xmlParserCtxtPtr ctxt)
1779: {
1780: xmlParserInputPtr input;
1781:
1782: if (ctxt == NULL) return;
1783:
1784: while ((input = inputPop(ctxt)) != NULL) { /* Non consuming */
1785: xmlFreeInputStream(input);
1786: }
1787: if (ctxt->spaceTab != NULL) xmlFree(ctxt->spaceTab);
1788: if (ctxt->nameTab != NULL) xmlFree((xmlChar * *)ctxt->nameTab);
1789: if (ctxt->nodeTab != NULL) xmlFree(ctxt->nodeTab);
1790: if (ctxt->nodeInfoTab != NULL) xmlFree(ctxt->nodeInfoTab);
1791: if (ctxt->inputTab != NULL) xmlFree(ctxt->inputTab);
1792: if (ctxt->version != NULL) xmlFree((char *) ctxt->version);
1793: if (ctxt->encoding != NULL) xmlFree((char *) ctxt->encoding);
1794: if (ctxt->extSubURI != NULL) xmlFree((char *) ctxt->extSubURI);
1795: if (ctxt->extSubSystem != NULL) xmlFree((char *) ctxt->extSubSystem);
1796: #ifdef LIBXML_SAX1_ENABLED
1797: if ((ctxt->sax != NULL) &&
1798: (ctxt->sax != (xmlSAXHandlerPtr) &xmlDefaultSAXHandler))
1799: #else
1800: if (ctxt->sax != NULL)
1801: #endif /* LIBXML_SAX1_ENABLED */
1802: xmlFree(ctxt->sax);
1803: if (ctxt->directory != NULL) xmlFree((char *) ctxt->directory);
1804: if (ctxt->vctxt.nodeTab != NULL) xmlFree(ctxt->vctxt.nodeTab);
1805: if (ctxt->atts != NULL) xmlFree((xmlChar * *)ctxt->atts);
1806: if (ctxt->dict != NULL) xmlDictFree(ctxt->dict);
1807: if (ctxt->nsTab != NULL) xmlFree((char *) ctxt->nsTab);
1808: if (ctxt->pushTab != NULL) xmlFree(ctxt->pushTab);
1809: if (ctxt->attallocs != NULL) xmlFree(ctxt->attallocs);
1810: if (ctxt->attsDefault != NULL)
1811: xmlHashFree(ctxt->attsDefault, (xmlHashDeallocator) xmlFree);
1812: if (ctxt->attsSpecial != NULL)
1813: xmlHashFree(ctxt->attsSpecial, NULL);
1814: if (ctxt->freeElems != NULL) {
1815: xmlNodePtr cur, next;
1816:
1817: cur = ctxt->freeElems;
1818: while (cur != NULL) {
1819: next = cur->next;
1820: xmlFree(cur);
1821: cur = next;
1822: }
1823: }
1824: if (ctxt->freeAttrs != NULL) {
1825: xmlAttrPtr cur, next;
1826:
1827: cur = ctxt->freeAttrs;
1828: while (cur != NULL) {
1829: next = cur->next;
1830: xmlFree(cur);
1831: cur = next;
1832: }
1833: }
1834: /*
1835: * cleanup the error strings
1836: */
1837: if (ctxt->lastError.message != NULL)
1838: xmlFree(ctxt->lastError.message);
1839: if (ctxt->lastError.file != NULL)
1840: xmlFree(ctxt->lastError.file);
1841: if (ctxt->lastError.str1 != NULL)
1842: xmlFree(ctxt->lastError.str1);
1843: if (ctxt->lastError.str2 != NULL)
1844: xmlFree(ctxt->lastError.str2);
1845: if (ctxt->lastError.str3 != NULL)
1846: xmlFree(ctxt->lastError.str3);
1847:
1848: #ifdef LIBXML_CATALOG_ENABLED
1849: if (ctxt->catalogs != NULL)
1850: xmlCatalogFreeLocal(ctxt->catalogs);
1851: #endif
1852: xmlFree(ctxt);
1853: }
1854:
1855: /**
1856: * xmlNewParserCtxt:
1857: *
1858: * Allocate and initialize a new parser context.
1859: *
1860: * Returns the xmlParserCtxtPtr or NULL
1861: */
1862:
1863: xmlParserCtxtPtr
1864: xmlNewParserCtxt(void)
1865: {
1866: xmlParserCtxtPtr ctxt;
1867:
1868: ctxt = (xmlParserCtxtPtr) xmlMalloc(sizeof(xmlParserCtxt));
1869: if (ctxt == NULL) {
1870: xmlErrMemory(NULL, "cannot allocate parser context\n");
1871: return(NULL);
1872: }
1873: memset(ctxt, 0, sizeof(xmlParserCtxt));
1874: if (xmlInitParserCtxt(ctxt) < 0) {
1875: xmlFreeParserCtxt(ctxt);
1876: return(NULL);
1877: }
1878: return(ctxt);
1879: }
1880:
1881: /************************************************************************
1882: * *
1883: * Handling of node informations *
1884: * *
1885: ************************************************************************/
1886:
1887: /**
1888: * xmlClearParserCtxt:
1889: * @ctxt: an XML parser context
1890: *
1891: * Clear (release owned resources) and reinitialize a parser context
1892: */
1893:
1894: void
1895: xmlClearParserCtxt(xmlParserCtxtPtr ctxt)
1896: {
1897: if (ctxt==NULL)
1898: return;
1899: xmlClearNodeInfoSeq(&ctxt->node_seq);
1900: xmlCtxtReset(ctxt);
1901: }
1902:
1903:
1904: /**
1905: * xmlParserFindNodeInfo:
1906: * @ctx: an XML parser context
1907: * @node: an XML node within the tree
1908: *
1909: * Find the parser node info struct for a given node
1910: *
1911: * Returns an xmlParserNodeInfo block pointer or NULL
1912: */
1913: const xmlParserNodeInfo *
1914: xmlParserFindNodeInfo(const xmlParserCtxtPtr ctx, const xmlNodePtr node)
1915: {
1916: unsigned long pos;
1917:
1918: if ((ctx == NULL) || (node == NULL))
1919: return (NULL);
1920: /* Find position where node should be at */
1921: pos = xmlParserFindNodeInfoIndex(&ctx->node_seq, node);
1922: if (pos < ctx->node_seq.length
1923: && ctx->node_seq.buffer[pos].node == node)
1924: return &ctx->node_seq.buffer[pos];
1925: else
1926: return NULL;
1927: }
1928:
1929:
1930: /**
1931: * xmlInitNodeInfoSeq:
1932: * @seq: a node info sequence pointer
1933: *
1934: * -- Initialize (set to initial state) node info sequence
1935: */
1936: void
1937: xmlInitNodeInfoSeq(xmlParserNodeInfoSeqPtr seq)
1938: {
1939: if (seq == NULL)
1940: return;
1941: seq->length = 0;
1942: seq->maximum = 0;
1943: seq->buffer = NULL;
1944: }
1945:
1946: /**
1947: * xmlClearNodeInfoSeq:
1948: * @seq: a node info sequence pointer
1949: *
1950: * -- Clear (release memory and reinitialize) node
1951: * info sequence
1952: */
1953: void
1954: xmlClearNodeInfoSeq(xmlParserNodeInfoSeqPtr seq)
1955: {
1956: if (seq == NULL)
1957: return;
1958: if (seq->buffer != NULL)
1959: xmlFree(seq->buffer);
1960: xmlInitNodeInfoSeq(seq);
1961: }
1962:
1963: /**
1964: * xmlParserFindNodeInfoIndex:
1965: * @seq: a node info sequence pointer
1966: * @node: an XML node pointer
1967: *
1968: *
1969: * xmlParserFindNodeInfoIndex : Find the index that the info record for
1970: * the given node is or should be at in a sorted sequence
1971: *
1972: * Returns a long indicating the position of the record
1973: */
1974: unsigned long
1975: xmlParserFindNodeInfoIndex(const xmlParserNodeInfoSeqPtr seq,
1976: const xmlNodePtr node)
1977: {
1978: unsigned long upper, lower, middle;
1979: int found = 0;
1980:
1981: if ((seq == NULL) || (node == NULL))
1982: return ((unsigned long) -1);
1983:
1984: /* Do a binary search for the key */
1985: lower = 1;
1986: upper = seq->length;
1987: middle = 0;
1988: while (lower <= upper && !found) {
1989: middle = lower + (upper - lower) / 2;
1990: if (node == seq->buffer[middle - 1].node)
1991: found = 1;
1992: else if (node < seq->buffer[middle - 1].node)
1993: upper = middle - 1;
1994: else
1995: lower = middle + 1;
1996: }
1997:
1998: /* Return position */
1999: if (middle == 0 || seq->buffer[middle - 1].node < node)
2000: return middle;
2001: else
2002: return middle - 1;
2003: }
2004:
2005:
2006: /**
2007: * xmlParserAddNodeInfo:
2008: * @ctxt: an XML parser context
2009: * @info: a node info sequence pointer
2010: *
2011: * Insert node info record into the sorted sequence
2012: */
2013: void
2014: xmlParserAddNodeInfo(xmlParserCtxtPtr ctxt,
2015: const xmlParserNodeInfoPtr info)
2016: {
2017: unsigned long pos;
2018:
2019: if ((ctxt == NULL) || (info == NULL)) return;
2020:
2021: /* Find pos and check to see if node is already in the sequence */
2022: pos = xmlParserFindNodeInfoIndex(&ctxt->node_seq, (xmlNodePtr)
2023: info->node);
2024:
2025: if ((pos < ctxt->node_seq.length) &&
2026: (ctxt->node_seq.buffer != NULL) &&
2027: (ctxt->node_seq.buffer[pos].node == info->node)) {
2028: ctxt->node_seq.buffer[pos] = *info;
2029: }
2030:
2031: /* Otherwise, we need to add new node to buffer */
2032: else {
2033: if (ctxt->node_seq.length + 1 > ctxt->node_seq.maximum) {
2034: xmlParserNodeInfo *tmp_buffer;
2035: unsigned int byte_size;
2036:
2037: if (ctxt->node_seq.maximum == 0)
2038: ctxt->node_seq.maximum = 2;
2039: byte_size = (sizeof(*ctxt->node_seq.buffer) *
2040: (2 * ctxt->node_seq.maximum));
2041:
2042: if (ctxt->node_seq.buffer == NULL)
2043: tmp_buffer = (xmlParserNodeInfo *) xmlMalloc(byte_size);
2044: else
2045: tmp_buffer =
2046: (xmlParserNodeInfo *) xmlRealloc(ctxt->node_seq.buffer,
2047: byte_size);
2048:
2049: if (tmp_buffer == NULL) {
2050: xmlErrMemory(ctxt, "failed to allocate buffer\n");
2051: return;
2052: }
2053: ctxt->node_seq.buffer = tmp_buffer;
2054: ctxt->node_seq.maximum *= 2;
2055: }
2056:
2057: /* If position is not at end, move elements out of the way */
2058: if (pos != ctxt->node_seq.length) {
2059: unsigned long i;
2060:
2061: for (i = ctxt->node_seq.length; i > pos; i--)
2062: ctxt->node_seq.buffer[i] = ctxt->node_seq.buffer[i - 1];
2063: }
2064:
2065: /* Copy element and increase length */
2066: ctxt->node_seq.buffer[pos] = *info;
2067: ctxt->node_seq.length++;
2068: }
2069: }
2070:
2071: /************************************************************************
2072: * *
2073: * Defaults settings *
2074: * *
2075: ************************************************************************/
2076: /**
2077: * xmlPedanticParserDefault:
2078: * @val: int 0 or 1
2079: *
2080: * Set and return the previous value for enabling pedantic warnings.
2081: *
2082: * Returns the last value for 0 for no substitution, 1 for substitution.
2083: */
2084:
2085: int
2086: xmlPedanticParserDefault(int val) {
2087: int old = xmlPedanticParserDefaultValue;
2088:
2089: xmlPedanticParserDefaultValue = val;
2090: return(old);
2091: }
2092:
2093: /**
2094: * xmlLineNumbersDefault:
2095: * @val: int 0 or 1
2096: *
2097: * Set and return the previous value for enabling line numbers in elements
2098: * contents. This may break on old application and is turned off by default.
2099: *
2100: * Returns the last value for 0 for no substitution, 1 for substitution.
2101: */
2102:
2103: int
2104: xmlLineNumbersDefault(int val) {
2105: int old = xmlLineNumbersDefaultValue;
2106:
2107: xmlLineNumbersDefaultValue = val;
2108: return(old);
2109: }
2110:
2111: /**
2112: * xmlSubstituteEntitiesDefault:
2113: * @val: int 0 or 1
2114: *
2115: * Set and return the previous value for default entity support.
2116: * Initially the parser always keep entity references instead of substituting
2117: * entity values in the output. This function has to be used to change the
2118: * default parser behavior
2119: * SAX::substituteEntities() has to be used for changing that on a file by
2120: * file basis.
2121: *
2122: * Returns the last value for 0 for no substitution, 1 for substitution.
2123: */
2124:
2125: int
2126: xmlSubstituteEntitiesDefault(int val) {
2127: int old = xmlSubstituteEntitiesDefaultValue;
2128:
2129: xmlSubstituteEntitiesDefaultValue = val;
2130: return(old);
2131: }
2132:
2133: /**
2134: * xmlKeepBlanksDefault:
2135: * @val: int 0 or 1
2136: *
2137: * Set and return the previous value for default blanks text nodes support.
2138: * The 1.x version of the parser used an heuristic to try to detect
2139: * ignorable white spaces. As a result the SAX callback was generating
2140: * xmlSAX2IgnorableWhitespace() callbacks instead of characters() one, and when
2141: * using the DOM output text nodes containing those blanks were not generated.
2142: * The 2.x and later version will switch to the XML standard way and
2143: * ignorableWhitespace() are only generated when running the parser in
2144: * validating mode and when the current element doesn't allow CDATA or
2145: * mixed content.
2146: * This function is provided as a way to force the standard behavior
2147: * on 1.X libs and to switch back to the old mode for compatibility when
2148: * running 1.X client code on 2.X . Upgrade of 1.X code should be done
2149: * by using xmlIsBlankNode() commodity function to detect the "empty"
2150: * nodes generated.
2151: * This value also affect autogeneration of indentation when saving code
2152: * if blanks sections are kept, indentation is not generated.
2153: *
2154: * Returns the last value for 0 for no substitution, 1 for substitution.
2155: */
2156:
2157: int
2158: xmlKeepBlanksDefault(int val) {
2159: int old = xmlKeepBlanksDefaultValue;
2160:
2161: xmlKeepBlanksDefaultValue = val;
2162: if (!val) xmlIndentTreeOutput = 1;
2163: return(old);
2164: }
2165:
2166: #define bottom_parserInternals
2167: #include "elfgcchack.h"
FreeBSD-CVSweb <freebsd-cvsweb@FreeBSD.org>