VirtualBox

source: vbox/trunk/src/libs/libxml2-2.9.14/HTMLparser.c@ 102239

Last change on this file since 102239 was 95312, checked in by vboxsync, 2 years ago

libs/{curl,libxml2}: OSE export fixes, bugref:8515
  • Property svn:eol-style set to native
File size: 208.4 KB
Line 
1/*
2 * HTMLparser.c : an HTML 4.0 non-verifying parser
3 *
4 * See Copyright for the status of this software.
5 *
6 * [email protected]
7 */
8
9#define IN_LIBXML
10#include "libxml.h"
11#ifdef LIBXML_HTML_ENABLED
12
13#include <string.h>
14#ifdef HAVE_CTYPE_H
15#include <ctype.h>
16#endif
17#ifdef HAVE_STDLIB_H
18#include <stdlib.h>
19#endif
20#ifdef HAVE_SYS_STAT_H
21#include <sys/stat.h>
22#endif
23#ifdef HAVE_FCNTL_H
24#include <fcntl.h>
25#endif
26#ifdef HAVE_UNISTD_H
27#include <unistd.h>
28#endif
29#ifdef LIBXML_ZLIB_ENABLED
30#include <zlib.h>
31#endif
32
33#include <libxml/xmlmemory.h>
34#include <libxml/tree.h>
35#include <libxml/parser.h>
36#include <libxml/parserInternals.h>
37#include <libxml/xmlerror.h>
38#include <libxml/HTMLparser.h>
39#include <libxml/HTMLtree.h>
40#include <libxml/entities.h>
41#include <libxml/encoding.h>
42#include <libxml/valid.h>
43#include <libxml/xmlIO.h>
44#include <libxml/globals.h>
45#include <libxml/uri.h>
46
47#include "buf.h"
48#include "enc.h"
49
50#define HTML_MAX_NAMELEN 1000
51#define HTML_PARSER_BIG_BUFFER_SIZE 1000
52#define HTML_PARSER_BUFFER_SIZE 100
53
54/* #define DEBUG */
55/* #define DEBUG_PUSH */
56
57static int htmlOmittedDefaultValue = 1;
58
59xmlChar * htmlDecodeEntities(htmlParserCtxtPtr ctxt, int len,
60 xmlChar end, xmlChar end2, xmlChar end3);
61static void htmlParseComment(htmlParserCtxtPtr ctxt);
62
63/************************************************************************
64 * *
65 * Some factorized error routines *
66 * *
67 ************************************************************************/
68
69/**
70 * htmlErrMemory:
71 * @ctxt: an HTML parser context
72 * @extra: extra information
73 *
74 * Handle a redefinition of attribute error
75 */
76static void
77htmlErrMemory(xmlParserCtxtPtr ctxt, const char *extra)
78{
79 if ((ctxt != NULL) && (ctxt->disableSAX != 0) &&
80 (ctxt->instate == XML_PARSER_EOF))
81 return;
82 if (ctxt != NULL) {
83 ctxt->errNo = XML_ERR_NO_MEMORY;
84 ctxt->instate = XML_PARSER_EOF;
85 ctxt->disableSAX = 1;
86 }
87 if (extra)
88 __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER,
89 XML_ERR_NO_MEMORY, XML_ERR_FATAL, NULL, 0, extra,
90 NULL, NULL, 0, 0,
91 "Memory allocation failed : %s\n", extra);
92 else
93 __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER,
94 XML_ERR_NO_MEMORY, XML_ERR_FATAL, NULL, 0, NULL,
95 NULL, NULL, 0, 0, "Memory allocation failed\n");
96}
97
98/**
99 * htmlParseErr:
100 * @ctxt: an HTML parser context
101 * @error: the error number
102 * @msg: the error message
103 * @str1: string infor
104 * @str2: string infor
105 *
106 * Handle a fatal parser error, i.e. violating Well-Formedness constraints
107 */
108static void LIBXML_ATTR_FORMAT(3,0)
109htmlParseErr(xmlParserCtxtPtr ctxt, xmlParserErrors error,
110 const char *msg, const xmlChar *str1, const xmlChar *str2)
111{
112 if ((ctxt != NULL) && (ctxt->disableSAX != 0) &&
113 (ctxt->instate == XML_PARSER_EOF))
114 return;
115 if (ctxt != NULL)
116 ctxt->errNo = error;
117 __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_HTML, error,
118 XML_ERR_ERROR, NULL, 0,
119 (const char *) str1, (const char *) str2,
120 NULL, 0, 0,
121 msg, str1, str2);
122 if (ctxt != NULL)
123 ctxt->wellFormed = 0;
124}
125
126/**
127 * htmlParseErrInt:
128 * @ctxt: an HTML parser context
129 * @error: the error number
130 * @msg: the error message
131 * @val: integer info
132 *
133 * Handle a fatal parser error, i.e. violating Well-Formedness constraints
134 */
135static void LIBXML_ATTR_FORMAT(3,0)
136htmlParseErrInt(xmlParserCtxtPtr ctxt, xmlParserErrors error,
137 const char *msg, int val)
138{
139 if ((ctxt != NULL) && (ctxt->disableSAX != 0) &&
140 (ctxt->instate == XML_PARSER_EOF))
141 return;
142 if (ctxt != NULL)
143 ctxt->errNo = error;
144 __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_HTML, error,
145 XML_ERR_ERROR, NULL, 0, NULL, NULL,
146 NULL, val, 0, msg, val);
147 if (ctxt != NULL)
148 ctxt->wellFormed = 0;
149}
150
151/************************************************************************
152 * *
153 * Parser stacks related functions and macros *
154 * *
155 ************************************************************************/
156
157/**
158 * htmlnamePush:
159 * @ctxt: an HTML parser context
160 * @value: the element name
161 *
162 * Pushes a new element name on top of the name stack
163 *
164 * Returns 0 in case of error, the index in the stack otherwise
165 */
166static int
167htmlnamePush(htmlParserCtxtPtr ctxt, const xmlChar * value)
168{
169 if ((ctxt->html < 3) && (xmlStrEqual(value, BAD_CAST "head")))
170 ctxt->html = 3;
171 if ((ctxt->html < 10) && (xmlStrEqual(value, BAD_CAST "body")))
172 ctxt->html = 10;
173 if (ctxt->nameNr >= ctxt->nameMax) {
174 ctxt->nameMax *= 2;
175 ctxt->nameTab = (const xmlChar * *)
176 xmlRealloc((xmlChar * *)ctxt->nameTab,
177 ctxt->nameMax *
178 sizeof(ctxt->nameTab[0]));
179 if (ctxt->nameTab == NULL) {
180 htmlErrMemory(ctxt, NULL);
181 return (0);
182 }
183 }
184 ctxt->nameTab[ctxt->nameNr] = value;
185 ctxt->name = value;
186 return (ctxt->nameNr++);
187}
188/**
189 * htmlnamePop:
190 * @ctxt: an HTML parser context
191 *
192 * Pops the top element name from the name stack
193 *
194 * Returns the name just removed
195 */
196static const xmlChar *
197htmlnamePop(htmlParserCtxtPtr ctxt)
198{
199 const xmlChar *ret;
200
201 if (ctxt->nameNr <= 0)
202 return (NULL);
203 ctxt->nameNr--;
204 if (ctxt->nameNr < 0)
205 return (NULL);
206 if (ctxt->nameNr > 0)
207 ctxt->name = ctxt->nameTab[ctxt->nameNr - 1];
208 else
209 ctxt->name = NULL;
210 ret = ctxt->nameTab[ctxt->nameNr];
211 ctxt->nameTab[ctxt->nameNr] = NULL;
212 return (ret);
213}
214
215/**
216 * htmlNodeInfoPush:
217 * @ctxt: an HTML parser context
218 * @value: the node info
219 *
220 * Pushes a new element name on top of the node info stack
221 *
222 * Returns 0 in case of error, the index in the stack otherwise
223 */
224static int
225htmlNodeInfoPush(htmlParserCtxtPtr ctxt, htmlParserNodeInfo *value)
226{
227 if (ctxt->nodeInfoNr >= ctxt->nodeInfoMax) {
228 if (ctxt->nodeInfoMax == 0)
229 ctxt->nodeInfoMax = 5;
230 ctxt->nodeInfoMax *= 2;
231 ctxt->nodeInfoTab = (htmlParserNodeInfo *)
232 xmlRealloc((htmlParserNodeInfo *)ctxt->nodeInfoTab,
233 ctxt->nodeInfoMax *
234 sizeof(ctxt->nodeInfoTab[0]));
235 if (ctxt->nodeInfoTab == NULL) {
236 htmlErrMemory(ctxt, NULL);
237 return (0);
238 }
239 }
240 ctxt->nodeInfoTab[ctxt->nodeInfoNr] = *value;
241 ctxt->nodeInfo = &ctxt->nodeInfoTab[ctxt->nodeInfoNr];
242 return (ctxt->nodeInfoNr++);
243}
244
245/**
246 * htmlNodeInfoPop:
247 * @ctxt: an HTML parser context
248 *
249 * Pops the top element name from the node info stack
250 *
251 * Returns 0 in case of error, the pointer to NodeInfo otherwise
252 */
253static htmlParserNodeInfo *
254htmlNodeInfoPop(htmlParserCtxtPtr ctxt)
255{
256 if (ctxt->nodeInfoNr <= 0)
257 return (NULL);
258 ctxt->nodeInfoNr--;
259 if (ctxt->nodeInfoNr < 0)
260 return (NULL);
261 if (ctxt->nodeInfoNr > 0)
262 ctxt->nodeInfo = &ctxt->nodeInfoTab[ctxt->nodeInfoNr - 1];
263 else
264 ctxt->nodeInfo = NULL;
265 return &ctxt->nodeInfoTab[ctxt->nodeInfoNr];
266}
267
268/*
269 * Macros for accessing the content. Those should be used only by the parser,
270 * and not exported.
271 *
272 * Dirty macros, i.e. one need to make assumption on the context to use them
273 *
274 * CUR_PTR return the current pointer to the xmlChar to be parsed.
275 * CUR returns the current xmlChar value, i.e. a 8 bit value if compiled
276 * in ISO-Latin or UTF-8, and the current 16 bit value if compiled
277 * in UNICODE mode. This should be used internally by the parser
278 * only to compare to ASCII values otherwise it would break when
279 * running with UTF-8 encoding.
280 * NXT(n) returns the n'th next xmlChar. Same as CUR is should be used only
281 * to compare on ASCII based substring.
282 * UPP(n) returns the n'th next xmlChar converted to uppercase. Same as CUR
283 * it should be used only to compare on ASCII based substring.
284 * SKIP(n) Skip n xmlChar, and must also be used only to skip ASCII defined
285 * strings without newlines within the parser.
286 *
287 * Clean macros, not dependent of an ASCII context, expect UTF-8 encoding
288 *
289 * CURRENT Returns the current char value, with the full decoding of
290 * UTF-8 if we are using this mode. It returns an int.
291 * NEXT Skip to the next character, this does the proper decoding
292 * in UTF-8 mode. It also pop-up unfinished entities on the fly.
293 * NEXTL(l) Skip the current unicode character of l xmlChars long.
294 * COPY(to) copy one char to *to, increment CUR_PTR and to accordingly
295 */
296
297#define UPPER (toupper(*ctxt->input->cur))
298
299#define SKIP(val) ctxt->input->cur += (val),ctxt->input->col+=(val)
300
301#define NXT(val) ctxt->input->cur[(val)]
302
303#define UPP(val) (toupper(ctxt->input->cur[(val)]))
304
305#define CUR_PTR ctxt->input->cur
306#define BASE_PTR ctxt->input->base
307
308#define SHRINK if ((ctxt->input->cur - ctxt->input->base > 2 * INPUT_CHUNK) && \
309 (ctxt->input->end - ctxt->input->cur < 2 * INPUT_CHUNK)) \
310 xmlParserInputShrink(ctxt->input)
311
312#define GROW if ((ctxt->progressive == 0) && \
313 (ctxt->input->end - ctxt->input->cur < INPUT_CHUNK)) \
314 xmlParserInputGrow(ctxt->input, INPUT_CHUNK)
315
316#define CURRENT ((int) (*ctxt->input->cur))
317
318#define SKIP_BLANKS htmlSkipBlankChars(ctxt)
319
320/* Imported from XML */
321
322/* #define CUR (ctxt->token ? ctxt->token : (int) (*ctxt->input->cur)) */
323#define CUR ((int) (*ctxt->input->cur))
324#define NEXT xmlNextChar(ctxt)
325
326#define RAW (ctxt->token ? -1 : (*ctxt->input->cur))
327
328
329#define NEXTL(l) do { \
330 if (*(ctxt->input->cur) == '\n') { \
331 ctxt->input->line++; ctxt->input->col = 1; \
332 } else ctxt->input->col++; \
333 ctxt->token = 0; ctxt->input->cur += l; \
334 } while (0)
335
336/************
337 \
338 if (*ctxt->input->cur == '%') xmlParserHandlePEReference(ctxt); \
339 if (*ctxt->input->cur == '&') xmlParserHandleReference(ctxt);
340 ************/
341
342#define CUR_CHAR(l) htmlCurrentChar(ctxt, &l)
343#define CUR_SCHAR(s, l) xmlStringCurrentChar(ctxt, s, &l)
344
345#define COPY_BUF(l,b,i,v) \
346 if (l == 1) b[i++] = (xmlChar) v; \
347 else i += xmlCopyChar(l,&b[i],v)
348
349/**
350 * htmlFindEncoding:
351 * @the HTML parser context
352 *
353 * Ty to find and encoding in the current data available in the input
354 * buffer this is needed to try to switch to the proper encoding when
355 * one face a character error.
356 * That's an heuristic, since it's operating outside of parsing it could
357 * try to use a meta which had been commented out, that's the reason it
358 * should only be used in case of error, not as a default.
359 *
360 * Returns an encoding string or NULL if not found, the string need to
361 * be freed
362 */
363static xmlChar *
364htmlFindEncoding(xmlParserCtxtPtr ctxt) {
365 const xmlChar *start, *cur, *end;
366
367 if ((ctxt == NULL) || (ctxt->input == NULL) ||
368 (ctxt->input->encoding != NULL) || (ctxt->input->buf == NULL) ||
369 (ctxt->input->buf->encoder != NULL))
370 return(NULL);
371 if ((ctxt->input->cur == NULL) || (ctxt->input->end == NULL))
372 return(NULL);
373
374 start = ctxt->input->cur;
375 end = ctxt->input->end;
376 /* we also expect the input buffer to be zero terminated */
377 if (*end != 0)
378 return(NULL);
379
380 cur = xmlStrcasestr(start, BAD_CAST "HTTP-EQUIV");
381 if (cur == NULL)
382 return(NULL);
383 cur = xmlStrcasestr(cur, BAD_CAST "CONTENT");
384 if (cur == NULL)
385 return(NULL);
386 cur = xmlStrcasestr(cur, BAD_CAST "CHARSET=");
387 if (cur == NULL)
388 return(NULL);
389 cur += 8;
390 start = cur;
391 while (((*cur >= 'A') && (*cur <= 'Z')) ||
392 ((*cur >= 'a') && (*cur <= 'z')) ||
393 ((*cur >= '0') && (*cur <= '9')) ||
394 (*cur == '-') || (*cur == '_') || (*cur == ':') || (*cur == '/'))
395 cur++;
396 if (cur == start)
397 return(NULL);
398 return(xmlStrndup(start, cur - start));
399}
400
401/**
402 * htmlCurrentChar:
403 * @ctxt: the HTML parser context
404 * @len: pointer to the length of the char read
405 *
406 * The current char value, if using UTF-8 this may actually span multiple
407 * bytes in the input buffer. Implement the end of line normalization:
408 * 2.11 End-of-Line Handling
409 * If the encoding is unspecified, in the case we find an ISO-Latin-1
410 * char, then the encoding converter is plugged in automatically.
411 *
412 * Returns the current char value and its length
413 */
414
415static int
416htmlCurrentChar(xmlParserCtxtPtr ctxt, int *len) {
417 const unsigned char *cur;
418 unsigned char c;
419 unsigned int val;
420
421 if (ctxt->instate == XML_PARSER_EOF)
422 return(0);
423
424 if (ctxt->token != 0) {
425 *len = 0;
426 return(ctxt->token);
427 }
428 if (ctxt->charset != XML_CHAR_ENCODING_UTF8) {
429 xmlChar * guess;
430 xmlCharEncodingHandlerPtr handler;
431
432 /*
433 * Assume it's a fixed length encoding (1) with
434 * a compatible encoding for the ASCII set, since
435 * HTML constructs only use < 128 chars
436 */
437 if ((int) *ctxt->input->cur < 0x80) {
438 *len = 1;
439 if ((*ctxt->input->cur == 0) &&
440 (ctxt->input->cur < ctxt->input->end)) {
441 htmlParseErrInt(ctxt, XML_ERR_INVALID_CHAR,
442 "Char 0x%X out of allowed range\n", 0);
443 return(' ');
444 }
445 return((int) *ctxt->input->cur);
446 }
447
448 /*
449 * Humm this is bad, do an automatic flow conversion
450 */
451 guess = htmlFindEncoding(ctxt);
452 if (guess == NULL) {
453 xmlSwitchEncoding(ctxt, XML_CHAR_ENCODING_8859_1);
454 } else {
455 if (ctxt->input->encoding != NULL)
456 xmlFree((xmlChar *) ctxt->input->encoding);
457 ctxt->input->encoding = guess;
458 handler = xmlFindCharEncodingHandler((const char *) guess);
459 if (handler != NULL) {
460 /*
461 * Don't use UTF-8 encoder which isn't required and
462 * can produce invalid UTF-8.
463 */
464 if (!xmlStrEqual(BAD_CAST handler->name, BAD_CAST "UTF-8"))
465 xmlSwitchToEncoding(ctxt, handler);
466 } else {
467 htmlParseErr(ctxt, XML_ERR_INVALID_ENCODING,
468 "Unsupported encoding %s", guess, NULL);
469 }
470 }
471 ctxt->charset = XML_CHAR_ENCODING_UTF8;
472 }
473
474 /*
475 * We are supposed to handle UTF8, check it's valid
476 * From rfc2044: encoding of the Unicode values on UTF-8:
477 *
478 * UCS-4 range (hex.) UTF-8 octet sequence (binary)
479 * 0000 0000-0000 007F 0xxxxxxx
480 * 0000 0080-0000 07FF 110xxxxx 10xxxxxx
481 * 0000 0800-0000 FFFF 1110xxxx 10xxxxxx 10xxxxxx
482 *
483 * Check for the 0x110000 limit too
484 */
485 cur = ctxt->input->cur;
486 c = *cur;
487 if (c & 0x80) {
488 if ((c & 0x40) == 0)
489 goto encoding_error;
490 if (cur[1] == 0) {
491 xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
492 cur = ctxt->input->cur;
493 }
494 if ((cur[1] & 0xc0) != 0x80)
495 goto encoding_error;
496 if ((c & 0xe0) == 0xe0) {
497
498 if (cur[2] == 0) {
499 xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
500 cur = ctxt->input->cur;
501 }
502 if ((cur[2] & 0xc0) != 0x80)
503 goto encoding_error;
504 if ((c & 0xf0) == 0xf0) {
505 if (cur[3] == 0) {
506 xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
507 cur = ctxt->input->cur;
508 }
509 if (((c & 0xf8) != 0xf0) ||
510 ((cur[3] & 0xc0) != 0x80))
511 goto encoding_error;
512 /* 4-byte code */
513 *len = 4;
514 val = (cur[0] & 0x7) << 18;
515 val |= (cur[1] & 0x3f) << 12;
516 val |= (cur[2] & 0x3f) << 6;
517 val |= cur[3] & 0x3f;
518 if (val < 0x10000)
519 goto encoding_error;
520 } else {
521 /* 3-byte code */
522 *len = 3;
523 val = (cur[0] & 0xf) << 12;
524 val |= (cur[1] & 0x3f) << 6;
525 val |= cur[2] & 0x3f;
526 if (val < 0x800)
527 goto encoding_error;
528 }
529 } else {
530 /* 2-byte code */
531 *len = 2;
532 val = (cur[0] & 0x1f) << 6;
533 val |= cur[1] & 0x3f;
534 if (val < 0x80)
535 goto encoding_error;
536 }
537 if (!IS_CHAR(val)) {
538 htmlParseErrInt(ctxt, XML_ERR_INVALID_CHAR,
539 "Char 0x%X out of allowed range\n", val);
540 }
541 return(val);
542 } else {
543 if ((*ctxt->input->cur == 0) &&
544 (ctxt->input->cur < ctxt->input->end)) {
545 htmlParseErrInt(ctxt, XML_ERR_INVALID_CHAR,
546 "Char 0x%X out of allowed range\n", 0);
547 *len = 1;
548 return(' ');
549 }
550 /* 1-byte code */
551 *len = 1;
552 return((int) *ctxt->input->cur);
553 }
554
555encoding_error:
556 /*
557 * If we detect an UTF8 error that probably mean that the
558 * input encoding didn't get properly advertised in the
559 * declaration header. Report the error and switch the encoding
560 * to ISO-Latin-1 (if you don't like this policy, just declare the
561 * encoding !)
562 */
563 {
564 char buffer[150];
565
566 if (ctxt->input->end - ctxt->input->cur >= 4) {
567 snprintf(buffer, 149, "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n",
568 ctxt->input->cur[0], ctxt->input->cur[1],
569 ctxt->input->cur[2], ctxt->input->cur[3]);
570 } else {
571 snprintf(buffer, 149, "Bytes: 0x%02X\n", ctxt->input->cur[0]);
572 }
573 htmlParseErr(ctxt, XML_ERR_INVALID_ENCODING,
574 "Input is not proper UTF-8, indicate encoding !\n",
575 BAD_CAST buffer, NULL);
576 }
577
578 /*
579 * Don't switch encodings twice. Note that if there's an encoder, we
580 * shouldn't receive invalid UTF-8 anyway.
581 *
582 * Note that if ctxt->input->buf == NULL, switching encodings is
583 * impossible, see Gitlab issue #34.
584 */
585 if ((ctxt->input->buf != NULL) &&
586 (ctxt->input->buf->encoder == NULL))
587 xmlSwitchEncoding(ctxt, XML_CHAR_ENCODING_8859_1);
588 *len = 1;
589 return((int) *ctxt->input->cur);
590}
591
592/**
593 * htmlSkipBlankChars:
594 * @ctxt: the HTML parser context
595 *
596 * skip all blanks character found at that point in the input streams.
597 *
598 * Returns the number of space chars skipped
599 */
600
601static int
602htmlSkipBlankChars(xmlParserCtxtPtr ctxt) {
603 int res = 0;
604
605 while (IS_BLANK_CH(*(ctxt->input->cur))) {
606 if ((*ctxt->input->cur == 0) &&
607 (xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0)) {
608 xmlPopInput(ctxt);
609 } else {
610 if (*(ctxt->input->cur) == '\n') {
611 ctxt->input->line++; ctxt->input->col = 1;
612 } else ctxt->input->col++;
613 ctxt->input->cur++;
614 if (*ctxt->input->cur == 0)
615 xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
616 }
617 if (res < INT_MAX)
618 res++;
619 }
620 return(res);
621}
622
623
624
625/************************************************************************
626 * *
627 * The list of HTML elements and their properties *
628 * *
629 ************************************************************************/
630
631/*
632 * Start Tag: 1 means the start tag can be omitted
633 * End Tag: 1 means the end tag can be omitted
634 * 2 means it's forbidden (empty elements)
635 * 3 means the tag is stylistic and should be closed easily
636 * Depr: this element is deprecated
637 * DTD: 1 means that this element is valid only in the Loose DTD
638 * 2 means that this element is valid only in the Frameset DTD
639 *
640 * Name,Start Tag,End Tag,Save End,Empty,Deprecated,DTD,inline,Description
641 , subElements , impliedsubelt , Attributes, userdata
642 */
643
644/* Definitions and a couple of vars for HTML Elements */
645
646#define FONTSTYLE "tt", "i", "b", "u", "s", "strike", "big", "small"
647#define NB_FONTSTYLE 8
648#define PHRASE "em", "strong", "dfn", "code", "samp", "kbd", "var", "cite", "abbr", "acronym"
649#define NB_PHRASE 10
650#define SPECIAL "a", "img", "applet", "embed", "object", "font", "basefont", "br", "script", "map", "q", "sub", "sup", "span", "bdo", "iframe"
651#define NB_SPECIAL 16
652#define INLINE FONTSTYLE, PHRASE, SPECIAL, FORMCTRL
653#define NB_INLINE NB_PCDATA + NB_FONTSTYLE + NB_PHRASE + NB_SPECIAL + NB_FORMCTRL
654#define BLOCK HEADING, LIST, "pre", "p", "dl", "div", "center", "noscript", "noframes", "blockquote", "form", "isindex", "hr", "table", "fieldset", "address"
655#define NB_BLOCK NB_HEADING + NB_LIST + 14
656#define FORMCTRL "input", "select", "textarea", "label", "button"
657#define NB_FORMCTRL 5
658#define PCDATA
659#define NB_PCDATA 0
660#define HEADING "h1", "h2", "h3", "h4", "h5", "h6"
661#define NB_HEADING 6
662#define LIST "ul", "ol", "dir", "menu"
663#define NB_LIST 4
664#define MODIFIER
665#define NB_MODIFIER 0
666#define FLOW BLOCK,INLINE
667#define NB_FLOW NB_BLOCK + NB_INLINE
668#define EMPTY NULL
669
670
671static const char* const html_flow[] = { FLOW, NULL } ;
672static const char* const html_inline[] = { INLINE, NULL } ;
673
674/* placeholders: elts with content but no subelements */
675static const char* const html_pcdata[] = { NULL } ;
676#define html_cdata html_pcdata
677
678
679/* ... and for HTML Attributes */
680
681#define COREATTRS "id", "class", "style", "title"
682#define NB_COREATTRS 4
683#define I18N "lang", "dir"
684#define NB_I18N 2
685#define EVENTS "onclick", "ondblclick", "onmousedown", "onmouseup", "onmouseover", "onmouseout", "onkeypress", "onkeydown", "onkeyup"
686#define NB_EVENTS 9
687#define ATTRS COREATTRS,I18N,EVENTS
688#define NB_ATTRS NB_NB_COREATTRS + NB_I18N + NB_EVENTS
689#define CELLHALIGN "align", "char", "charoff"
690#define NB_CELLHALIGN 3
691#define CELLVALIGN "valign"
692#define NB_CELLVALIGN 1
693
694static const char* const html_attrs[] = { ATTRS, NULL } ;
695static const char* const core_i18n_attrs[] = { COREATTRS, I18N, NULL } ;
696static const char* const core_attrs[] = { COREATTRS, NULL } ;
697static const char* const i18n_attrs[] = { I18N, NULL } ;
698
699
700/* Other declarations that should go inline ... */
701static const char* const a_attrs[] = { ATTRS, "charset", "type", "name",
702 "href", "hreflang", "rel", "rev", "accesskey", "shape", "coords",
703 "tabindex", "onfocus", "onblur", NULL } ;
704static const char* const target_attr[] = { "target", NULL } ;
705static const char* const rows_cols_attr[] = { "rows", "cols", NULL } ;
706static const char* const alt_attr[] = { "alt", NULL } ;
707static const char* const src_alt_attrs[] = { "src", "alt", NULL } ;
708static const char* const href_attrs[] = { "href", NULL } ;
709static const char* const clear_attrs[] = { "clear", NULL } ;
710static const char* const inline_p[] = { INLINE, "p", NULL } ;
711
712static const char* const flow_param[] = { FLOW, "param", NULL } ;
713static const char* const applet_attrs[] = { COREATTRS , "codebase",
714 "archive", "alt", "name", "height", "width", "align",
715 "hspace", "vspace", NULL } ;
716static const char* const area_attrs[] = { "shape", "coords", "href", "nohref",
717 "tabindex", "accesskey", "onfocus", "onblur", NULL } ;
718static const char* const basefont_attrs[] =
719 { "id", "size", "color", "face", NULL } ;
720static const char* const quote_attrs[] = { ATTRS, "cite", NULL } ;
721static const char* const body_contents[] = { FLOW, "ins", "del", NULL } ;
722static const char* const body_attrs[] = { ATTRS, "onload", "onunload", NULL } ;
723static const char* const body_depr[] = { "background", "bgcolor", "text",
724 "link", "vlink", "alink", NULL } ;
725static const char* const button_attrs[] = { ATTRS, "name", "value", "type",
726 "disabled", "tabindex", "accesskey", "onfocus", "onblur", NULL } ;
727
728
729static const char* const col_attrs[] = { ATTRS, "span", "width", CELLHALIGN, CELLVALIGN, NULL } ;
730static const char* const col_elt[] = { "col", NULL } ;
731static const char* const edit_attrs[] = { ATTRS, "datetime", "cite", NULL } ;
732static const char* const compact_attrs[] = { ATTRS, "compact", NULL } ;
733static const char* const dl_contents[] = { "dt", "dd", NULL } ;
734static const char* const compact_attr[] = { "compact", NULL } ;
735static const char* const label_attr[] = { "label", NULL } ;
736static const char* const fieldset_contents[] = { FLOW, "legend" } ;
737static const char* const font_attrs[] = { COREATTRS, I18N, "size", "color", "face" , NULL } ;
738static const char* const form_contents[] = { HEADING, LIST, INLINE, "pre", "p", "div", "center", "noscript", "noframes", "blockquote", "isindex", "hr", "table", "fieldset", "address", NULL } ;
739static const char* const form_attrs[] = { ATTRS, "method", "enctype", "accept", "name", "onsubmit", "onreset", "accept-charset", NULL } ;
740static const char* const frame_attrs[] = { COREATTRS, "longdesc", "name", "src", "frameborder", "marginwidth", "marginheight", "noresize", "scrolling" , NULL } ;
741static const char* const frameset_attrs[] = { COREATTRS, "rows", "cols", "onload", "onunload", NULL } ;
742static const char* const frameset_contents[] = { "frameset", "frame", "noframes", NULL } ;
743static const char* const head_attrs[] = { I18N, "profile", NULL } ;
744static const char* const head_contents[] = { "title", "isindex", "base", "script", "style", "meta", "link", "object", NULL } ;
745static const char* const hr_depr[] = { "align", "noshade", "size", "width", NULL } ;
746static const char* const version_attr[] = { "version", NULL } ;
747static const char* const html_content[] = { "head", "body", "frameset", NULL } ;
748static const char* const iframe_attrs[] = { COREATTRS, "longdesc", "name", "src", "frameborder", "marginwidth", "marginheight", "scrolling", "align", "height", "width", NULL } ;
749static const char* const img_attrs[] = { ATTRS, "longdesc", "name", "height", "width", "usemap", "ismap", NULL } ;
750static const char* const embed_attrs[] = { COREATTRS, "align", "alt", "border", "code", "codebase", "frameborder", "height", "hidden", "hspace", "name", "palette", "pluginspace", "pluginurl", "src", "type", "units", "vspace", "width", NULL } ;
751static const char* const input_attrs[] = { ATTRS, "type", "name", "value", "checked", "disabled", "readonly", "size", "maxlength", "src", "alt", "usemap", "ismap", "tabindex", "accesskey", "onfocus", "onblur", "onselect", "onchange", "accept", NULL } ;
752static const char* const prompt_attrs[] = { COREATTRS, I18N, "prompt", NULL } ;
753static const char* const label_attrs[] = { ATTRS, "for", "accesskey", "onfocus", "onblur", NULL } ;
754static const char* const legend_attrs[] = { ATTRS, "accesskey", NULL } ;
755static const char* const align_attr[] = { "align", NULL } ;
756static const char* const link_attrs[] = { ATTRS, "charset", "href", "hreflang", "type", "rel", "rev", "media", NULL } ;
757static const char* const map_contents[] = { BLOCK, "area", NULL } ;
758static const char* const name_attr[] = { "name", NULL } ;
759static const char* const action_attr[] = { "action", NULL } ;
760static const char* const blockli_elt[] = { BLOCK, "li", NULL } ;
761static const char* const meta_attrs[] = { I18N, "http-equiv", "name", "scheme", "charset", NULL } ;
762static const char* const content_attr[] = { "content", NULL } ;
763static const char* const type_attr[] = { "type", NULL } ;
764static const char* const noframes_content[] = { "body", FLOW MODIFIER, NULL } ;
765static const char* const object_contents[] = { FLOW, "param", NULL } ;
766static const char* const object_attrs[] = { ATTRS, "declare", "classid", "codebase", "data", "type", "codetype", "archive", "standby", "height", "width", "usemap", "name", "tabindex", NULL } ;
767static const char* const object_depr[] = { "align", "border", "hspace", "vspace", NULL } ;
768static const char* const ol_attrs[] = { "type", "compact", "start", NULL} ;
769static const char* const option_elt[] = { "option", NULL } ;
770static const char* const optgroup_attrs[] = { ATTRS, "disabled", NULL } ;
771static const char* const option_attrs[] = { ATTRS, "disabled", "label", "selected", "value", NULL } ;
772static const char* const param_attrs[] = { "id", "value", "valuetype", "type", NULL } ;
773static const char* const width_attr[] = { "width", NULL } ;
774static const char* const pre_content[] = { PHRASE, "tt", "i", "b", "u", "s", "strike", "a", "br", "script", "map", "q", "span", "bdo", "iframe", NULL } ;
775static const char* const script_attrs[] = { "charset", "src", "defer", "event", "for", NULL } ;
776static const char* const language_attr[] = { "language", NULL } ;
777static const char* const select_content[] = { "optgroup", "option", NULL } ;
778static const char* const select_attrs[] = { ATTRS, "name", "size", "multiple", "disabled", "tabindex", "onfocus", "onblur", "onchange", NULL } ;
779static const char* const style_attrs[] = { I18N, "media", "title", NULL } ;
780static const char* const table_attrs[] = { ATTRS, "summary", "width", "border", "frame", "rules", "cellspacing", "cellpadding", "datapagesize", NULL } ;
781static const char* const table_depr[] = { "align", "bgcolor", NULL } ;
782static const char* const table_contents[] = { "caption", "col", "colgroup", "thead", "tfoot", "tbody", "tr", NULL} ;
783static const char* const tr_elt[] = { "tr", NULL } ;
784static const char* const talign_attrs[] = { ATTRS, CELLHALIGN, CELLVALIGN, NULL} ;
785static const char* const th_td_depr[] = { "nowrap", "bgcolor", "width", "height", NULL } ;
786static const char* const th_td_attr[] = { ATTRS, "abbr", "axis", "headers", "scope", "rowspan", "colspan", CELLHALIGN, CELLVALIGN, NULL } ;
787static const char* const textarea_attrs[] = { ATTRS, "name", "disabled", "readonly", "tabindex", "accesskey", "onfocus", "onblur", "onselect", "onchange", NULL } ;
788static const char* const tr_contents[] = { "th", "td", NULL } ;
789static const char* const bgcolor_attr[] = { "bgcolor", NULL } ;
790static const char* const li_elt[] = { "li", NULL } ;
791static const char* const ul_depr[] = { "type", "compact", NULL} ;
792static const char* const dir_attr[] = { "dir", NULL} ;
793
794#define DECL (const char**)
795
796static const htmlElemDesc
797html40ElementTable[] = {
798{ "a", 0, 0, 0, 0, 0, 0, 1, "anchor ",
799 DECL html_inline , NULL , DECL a_attrs , DECL target_attr, NULL
800},
801{ "abbr", 0, 0, 0, 0, 0, 0, 1, "abbreviated form",
802 DECL html_inline , NULL , DECL html_attrs, NULL, NULL
803},
804{ "acronym", 0, 0, 0, 0, 0, 0, 1, "",
805 DECL html_inline , NULL , DECL html_attrs, NULL, NULL
806},
807{ "address", 0, 0, 0, 0, 0, 0, 0, "information on author ",
808 DECL inline_p , NULL , DECL html_attrs, NULL, NULL
809},
810{ "applet", 0, 0, 0, 0, 1, 1, 2, "java applet ",
811 DECL flow_param , NULL , NULL , DECL applet_attrs, NULL
812},
813{ "area", 0, 2, 2, 1, 0, 0, 0, "client-side image map area ",
814 EMPTY , NULL , DECL area_attrs , DECL target_attr, DECL alt_attr
815},
816{ "b", 0, 3, 0, 0, 0, 0, 1, "bold text style",
817 DECL html_inline , NULL , DECL html_attrs, NULL, NULL
818},
819{ "base", 0, 2, 2, 1, 0, 0, 0, "document base uri ",
820 EMPTY , NULL , NULL , DECL target_attr, DECL href_attrs
821},
822{ "basefont", 0, 2, 2, 1, 1, 1, 1, "base font size " ,
823 EMPTY , NULL , NULL, DECL basefont_attrs, NULL
824},
825{ "bdo", 0, 0, 0, 0, 0, 0, 1, "i18n bidi over-ride ",
826 DECL html_inline , NULL , DECL core_i18n_attrs, NULL, DECL dir_attr
827},
828{ "big", 0, 3, 0, 0, 0, 0, 1, "large text style",
829 DECL html_inline , NULL , DECL html_attrs, NULL, NULL
830},
831{ "blockquote", 0, 0, 0, 0, 0, 0, 0, "long quotation ",
832 DECL html_flow , NULL , DECL quote_attrs , NULL, NULL
833},
834{ "body", 1, 1, 0, 0, 0, 0, 0, "document body ",
835 DECL body_contents , "div" , DECL body_attrs, DECL body_depr, NULL
836},
837{ "br", 0, 2, 2, 1, 0, 0, 1, "forced line break ",
838 EMPTY , NULL , DECL core_attrs, DECL clear_attrs , NULL
839},
840{ "button", 0, 0, 0, 0, 0, 0, 2, "push button ",
841 DECL html_flow MODIFIER , NULL , DECL button_attrs, NULL, NULL
842},
843{ "caption", 0, 0, 0, 0, 0, 0, 0, "table caption ",
844 DECL html_inline , NULL , DECL html_attrs, NULL, NULL
845},
846{ "center", 0, 3, 0, 0, 1, 1, 0, "shorthand for div align=center ",
847 DECL html_flow , NULL , NULL, DECL html_attrs, NULL
848},
849{ "cite", 0, 0, 0, 0, 0, 0, 1, "citation",
850 DECL html_inline , NULL , DECL html_attrs, NULL, NULL
851},
852{ "code", 0, 0, 0, 0, 0, 0, 1, "computer code fragment",
853 DECL html_inline , NULL , DECL html_attrs, NULL, NULL
854},
855{ "col", 0, 2, 2, 1, 0, 0, 0, "table column ",
856 EMPTY , NULL , DECL col_attrs , NULL, NULL
857},
858{ "colgroup", 0, 1, 0, 0, 0, 0, 0, "table column group ",
859 DECL col_elt , "col" , DECL col_attrs , NULL, NULL
860},
861{ "dd", 0, 1, 0, 0, 0, 0, 0, "definition description ",
862 DECL html_flow , NULL , DECL html_attrs, NULL, NULL
863},
864{ "del", 0, 0, 0, 0, 0, 0, 2, "deleted text ",
865 DECL html_flow , NULL , DECL edit_attrs , NULL, NULL
866},
867{ "dfn", 0, 0, 0, 0, 0, 0, 1, "instance definition",
868 DECL html_inline , NULL , DECL html_attrs, NULL, NULL
869},
870{ "dir", 0, 0, 0, 0, 1, 1, 0, "directory list",
871 DECL blockli_elt, "li" , NULL, DECL compact_attrs, NULL
872},
873{ "div", 0, 0, 0, 0, 0, 0, 0, "generic language/style container",
874 DECL html_flow, NULL, DECL html_attrs, DECL align_attr, NULL
875},
876{ "dl", 0, 0, 0, 0, 0, 0, 0, "definition list ",
877 DECL dl_contents , "dd" , DECL html_attrs, DECL compact_attr, NULL
878},
879{ "dt", 0, 1, 0, 0, 0, 0, 0, "definition term ",
880 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
881},
882{ "em", 0, 3, 0, 0, 0, 0, 1, "emphasis",
883 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
884},
885{ "embed", 0, 1, 0, 0, 1, 1, 1, "generic embedded object ",
886 EMPTY, NULL, DECL embed_attrs, NULL, NULL
887},
888{ "fieldset", 0, 0, 0, 0, 0, 0, 0, "form control group ",
889 DECL fieldset_contents , NULL, DECL html_attrs, NULL, NULL
890},
891{ "font", 0, 3, 0, 0, 1, 1, 1, "local change to font ",
892 DECL html_inline, NULL, NULL, DECL font_attrs, NULL
893},
894{ "form", 0, 0, 0, 0, 0, 0, 0, "interactive form ",
895 DECL form_contents, "fieldset", DECL form_attrs , DECL target_attr, DECL action_attr
896},
897{ "frame", 0, 2, 2, 1, 0, 2, 0, "subwindow " ,
898 EMPTY, NULL, NULL, DECL frame_attrs, NULL
899},
900{ "frameset", 0, 0, 0, 0, 0, 2, 0, "window subdivision" ,
901 DECL frameset_contents, "noframes" , NULL , DECL frameset_attrs, NULL
902},
903{ "h1", 0, 0, 0, 0, 0, 0, 0, "heading ",
904 DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
905},
906{ "h2", 0, 0, 0, 0, 0, 0, 0, "heading ",
907 DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
908},
909{ "h3", 0, 0, 0, 0, 0, 0, 0, "heading ",
910 DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
911},
912{ "h4", 0, 0, 0, 0, 0, 0, 0, "heading ",
913 DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
914},
915{ "h5", 0, 0, 0, 0, 0, 0, 0, "heading ",
916 DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
917},
918{ "h6", 0, 0, 0, 0, 0, 0, 0, "heading ",
919 DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
920},
921{ "head", 1, 1, 0, 0, 0, 0, 0, "document head ",
922 DECL head_contents, NULL, DECL head_attrs, NULL, NULL
923},
924{ "hr", 0, 2, 2, 1, 0, 0, 0, "horizontal rule " ,
925 EMPTY, NULL, DECL html_attrs, DECL hr_depr, NULL
926},
927{ "html", 1, 1, 0, 0, 0, 0, 0, "document root element ",
928 DECL html_content , NULL , DECL i18n_attrs, DECL version_attr, NULL
929},
930{ "i", 0, 3, 0, 0, 0, 0, 1, "italic text style",
931 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
932},
933{ "iframe", 0, 0, 0, 0, 0, 1, 2, "inline subwindow ",
934 DECL html_flow, NULL, NULL, DECL iframe_attrs, NULL
935},
936{ "img", 0, 2, 2, 1, 0, 0, 1, "embedded image ",
937 EMPTY, NULL, DECL img_attrs, DECL align_attr, DECL src_alt_attrs
938},
939{ "input", 0, 2, 2, 1, 0, 0, 1, "form control ",
940 EMPTY, NULL, DECL input_attrs , DECL align_attr, NULL
941},
942{ "ins", 0, 0, 0, 0, 0, 0, 2, "inserted text",
943 DECL html_flow, NULL, DECL edit_attrs, NULL, NULL
944},
945{ "isindex", 0, 2, 2, 1, 1, 1, 0, "single line prompt ",
946 EMPTY, NULL, NULL, DECL prompt_attrs, NULL
947},
948{ "kbd", 0, 0, 0, 0, 0, 0, 1, "text to be entered by the user",
949 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
950},
951{ "label", 0, 0, 0, 0, 0, 0, 1, "form field label text ",
952 DECL html_inline MODIFIER, NULL, DECL label_attrs , NULL, NULL
953},
954{ "legend", 0, 0, 0, 0, 0, 0, 0, "fieldset legend ",
955 DECL html_inline, NULL, DECL legend_attrs , DECL align_attr, NULL
956},
957{ "li", 0, 1, 1, 0, 0, 0, 0, "list item ",
958 DECL html_flow, NULL, DECL html_attrs, NULL, NULL
959},
960{ "link", 0, 2, 2, 1, 0, 0, 0, "a media-independent link ",
961 EMPTY, NULL, DECL link_attrs, DECL target_attr, NULL
962},
963{ "map", 0, 0, 0, 0, 0, 0, 2, "client-side image map ",
964 DECL map_contents , NULL, DECL html_attrs , NULL, DECL name_attr
965},
966{ "menu", 0, 0, 0, 0, 1, 1, 0, "menu list ",
967 DECL blockli_elt , NULL, NULL, DECL compact_attrs, NULL
968},
969{ "meta", 0, 2, 2, 1, 0, 0, 0, "generic metainformation ",
970 EMPTY, NULL, DECL meta_attrs , NULL , DECL content_attr
971},
972{ "noframes", 0, 0, 0, 0, 0, 2, 0, "alternate content container for non frame-based rendering ",
973 DECL noframes_content, "body" , DECL html_attrs, NULL, NULL
974},
975{ "noscript", 0, 0, 0, 0, 0, 0, 0, "alternate content container for non script-based rendering ",
976 DECL html_flow, "div", DECL html_attrs, NULL, NULL
977},
978{ "object", 0, 0, 0, 0, 0, 0, 2, "generic embedded object ",
979 DECL object_contents , "div" , DECL object_attrs, DECL object_depr, NULL
980},
981{ "ol", 0, 0, 0, 0, 0, 0, 0, "ordered list ",
982 DECL li_elt , "li" , DECL html_attrs, DECL ol_attrs, NULL
983},
984{ "optgroup", 0, 0, 0, 0, 0, 0, 0, "option group ",
985 DECL option_elt , "option", DECL optgroup_attrs, NULL, DECL label_attr
986},
987{ "option", 0, 1, 0, 0, 0, 0, 0, "selectable choice " ,
988 DECL html_pcdata, NULL, DECL option_attrs, NULL, NULL
989},
990{ "p", 0, 1, 0, 0, 0, 0, 0, "paragraph ",
991 DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
992},
993{ "param", 0, 2, 2, 1, 0, 0, 0, "named property value ",
994 EMPTY, NULL, DECL param_attrs, NULL, DECL name_attr
995},
996{ "pre", 0, 0, 0, 0, 0, 0, 0, "preformatted text ",
997 DECL pre_content, NULL, DECL html_attrs, DECL width_attr, NULL
998},
999{ "q", 0, 0, 0, 0, 0, 0, 1, "short inline quotation ",
1000 DECL html_inline, NULL, DECL quote_attrs, NULL, NULL
1001},
1002{ "s", 0, 3, 0, 0, 1, 1, 1, "strike-through text style",
1003 DECL html_inline, NULL, NULL, DECL html_attrs, NULL
1004},
1005{ "samp", 0, 0, 0, 0, 0, 0, 1, "sample program output, scripts, etc.",
1006 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
1007},
1008{ "script", 0, 0, 0, 0, 0, 0, 2, "script statements ",
1009 DECL html_cdata, NULL, DECL script_attrs, DECL language_attr, DECL type_attr
1010},
1011{ "select", 0, 0, 0, 0, 0, 0, 1, "option selector ",
1012 DECL select_content, NULL, DECL select_attrs, NULL, NULL
1013},
1014{ "small", 0, 3, 0, 0, 0, 0, 1, "small text style",
1015 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
1016},
1017{ "span", 0, 0, 0, 0, 0, 0, 1, "generic language/style container ",
1018 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
1019},
1020{ "strike", 0, 3, 0, 0, 1, 1, 1, "strike-through text",
1021 DECL html_inline, NULL, NULL, DECL html_attrs, NULL
1022},
1023{ "strong", 0, 3, 0, 0, 0, 0, 1, "strong emphasis",
1024 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
1025},
1026{ "style", 0, 0, 0, 0, 0, 0, 0, "style info ",
1027 DECL html_cdata, NULL, DECL style_attrs, NULL, DECL type_attr
1028},
1029{ "sub", 0, 3, 0, 0, 0, 0, 1, "subscript",
1030 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
1031},
1032{ "sup", 0, 3, 0, 0, 0, 0, 1, "superscript ",
1033 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
1034},
1035{ "table", 0, 0, 0, 0, 0, 0, 0, "",
1036 DECL table_contents , "tr" , DECL table_attrs , DECL table_depr, NULL
1037},
1038{ "tbody", 1, 0, 0, 0, 0, 0, 0, "table body ",
1039 DECL tr_elt , "tr" , DECL talign_attrs, NULL, NULL
1040},
1041{ "td", 0, 0, 0, 0, 0, 0, 0, "table data cell",
1042 DECL html_flow, NULL, DECL th_td_attr, DECL th_td_depr, NULL
1043},
1044{ "textarea", 0, 0, 0, 0, 0, 0, 1, "multi-line text field ",
1045 DECL html_pcdata, NULL, DECL textarea_attrs, NULL, DECL rows_cols_attr
1046},
1047{ "tfoot", 0, 1, 0, 0, 0, 0, 0, "table footer ",
1048 DECL tr_elt , "tr" , DECL talign_attrs, NULL, NULL
1049},
1050{ "th", 0, 1, 0, 0, 0, 0, 0, "table header cell",
1051 DECL html_flow, NULL, DECL th_td_attr, DECL th_td_depr, NULL
1052},
1053{ "thead", 0, 1, 0, 0, 0, 0, 0, "table header ",
1054 DECL tr_elt , "tr" , DECL talign_attrs, NULL, NULL
1055},
1056{ "title", 0, 0, 0, 0, 0, 0, 0, "document title ",
1057 DECL html_pcdata, NULL, DECL i18n_attrs, NULL, NULL
1058},
1059{ "tr", 0, 0, 0, 0, 0, 0, 0, "table row ",
1060 DECL tr_contents , "td" , DECL talign_attrs, DECL bgcolor_attr, NULL
1061},
1062{ "tt", 0, 3, 0, 0, 0, 0, 1, "teletype or monospaced text style",
1063 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
1064},
1065{ "u", 0, 3, 0, 0, 1, 1, 1, "underlined text style",
1066 DECL html_inline, NULL, NULL, DECL html_attrs, NULL
1067},
1068{ "ul", 0, 0, 0, 0, 0, 0, 0, "unordered list ",
1069 DECL li_elt , "li" , DECL html_attrs, DECL ul_depr, NULL
1070},
1071{ "var", 0, 0, 0, 0, 0, 0, 1, "instance of a variable or program argument",
1072 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
1073}
1074};
1075
1076typedef struct {
1077 const char *oldTag;
1078 const char *newTag;
1079} htmlStartCloseEntry;
1080
1081/*
1082 * start tags that imply the end of current element
1083 */
1084static const htmlStartCloseEntry htmlStartClose[] = {
1085 { "a", "a" },
1086 { "a", "fieldset" },
1087 { "a", "table" },
1088 { "a", "td" },
1089 { "a", "th" },
1090 { "address", "dd" },
1091 { "address", "dl" },
1092 { "address", "dt" },
1093 { "address", "form" },
1094 { "address", "li" },
1095 { "address", "ul" },
1096 { "b", "center" },
1097 { "b", "p" },
1098 { "b", "td" },
1099 { "b", "th" },
1100 { "big", "p" },
1101 { "caption", "col" },
1102 { "caption", "colgroup" },
1103 { "caption", "tbody" },
1104 { "caption", "tfoot" },
1105 { "caption", "thead" },
1106 { "caption", "tr" },
1107 { "col", "col" },
1108 { "col", "colgroup" },
1109 { "col", "tbody" },
1110 { "col", "tfoot" },
1111 { "col", "thead" },
1112 { "col", "tr" },
1113 { "colgroup", "colgroup" },
1114 { "colgroup", "tbody" },
1115 { "colgroup", "tfoot" },
1116 { "colgroup", "thead" },
1117 { "colgroup", "tr" },
1118 { "dd", "dt" },
1119 { "dir", "dd" },
1120 { "dir", "dl" },
1121 { "dir", "dt" },
1122 { "dir", "form" },
1123 { "dir", "ul" },
1124 { "dl", "form" },
1125 { "dl", "li" },
1126 { "dt", "dd" },
1127 { "dt", "dl" },
1128 { "font", "center" },
1129 { "font", "td" },
1130 { "font", "th" },
1131 { "form", "form" },
1132 { "h1", "fieldset" },
1133 { "h1", "form" },
1134 { "h1", "li" },
1135 { "h1", "p" },
1136 { "h1", "table" },
1137 { "h2", "fieldset" },
1138 { "h2", "form" },
1139 { "h2", "li" },
1140 { "h2", "p" },
1141 { "h2", "table" },
1142 { "h3", "fieldset" },
1143 { "h3", "form" },
1144 { "h3", "li" },
1145 { "h3", "p" },
1146 { "h3", "table" },
1147 { "h4", "fieldset" },
1148 { "h4", "form" },
1149 { "h4", "li" },
1150 { "h4", "p" },
1151 { "h4", "table" },
1152 { "h5", "fieldset" },
1153 { "h5", "form" },
1154 { "h5", "li" },
1155 { "h5", "p" },
1156 { "h5", "table" },
1157 { "h6", "fieldset" },
1158 { "h6", "form" },
1159 { "h6", "li" },
1160 { "h6", "p" },
1161 { "h6", "table" },
1162 { "head", "a" },
1163 { "head", "abbr" },
1164 { "head", "acronym" },
1165 { "head", "address" },
1166 { "head", "b" },
1167 { "head", "bdo" },
1168 { "head", "big" },
1169 { "head", "blockquote" },
1170 { "head", "body" },
1171 { "head", "br" },
1172 { "head", "center" },
1173 { "head", "cite" },
1174 { "head", "code" },
1175 { "head", "dd" },
1176 { "head", "dfn" },
1177 { "head", "dir" },
1178 { "head", "div" },
1179 { "head", "dl" },
1180 { "head", "dt" },
1181 { "head", "em" },
1182 { "head", "fieldset" },
1183 { "head", "font" },
1184 { "head", "form" },
1185 { "head", "frameset" },
1186 { "head", "h1" },
1187 { "head", "h2" },
1188 { "head", "h3" },
1189 { "head", "h4" },
1190 { "head", "h5" },
1191 { "head", "h6" },
1192 { "head", "hr" },
1193 { "head", "i" },
1194 { "head", "iframe" },
1195 { "head", "img" },
1196 { "head", "kbd" },
1197 { "head", "li" },
1198 { "head", "listing" },
1199 { "head", "map" },
1200 { "head", "menu" },
1201 { "head", "ol" },
1202 { "head", "p" },
1203 { "head", "pre" },
1204 { "head", "q" },
1205 { "head", "s" },
1206 { "head", "samp" },
1207 { "head", "small" },
1208 { "head", "span" },
1209 { "head", "strike" },
1210 { "head", "strong" },
1211 { "head", "sub" },
1212 { "head", "sup" },
1213 { "head", "table" },
1214 { "head", "tt" },
1215 { "head", "u" },
1216 { "head", "ul" },
1217 { "head", "var" },
1218 { "head", "xmp" },
1219 { "hr", "form" },
1220 { "i", "center" },
1221 { "i", "p" },
1222 { "i", "td" },
1223 { "i", "th" },
1224 { "legend", "fieldset" },
1225 { "li", "li" },
1226 { "link", "body" },
1227 { "link", "frameset" },
1228 { "listing", "dd" },
1229 { "listing", "dl" },
1230 { "listing", "dt" },
1231 { "listing", "fieldset" },
1232 { "listing", "form" },
1233 { "listing", "li" },
1234 { "listing", "table" },
1235 { "listing", "ul" },
1236 { "menu", "dd" },
1237 { "menu", "dl" },
1238 { "menu", "dt" },
1239 { "menu", "form" },
1240 { "menu", "ul" },
1241 { "ol", "form" },
1242 { "ol", "ul" },
1243 { "option", "optgroup" },
1244 { "option", "option" },
1245 { "p", "address" },
1246 { "p", "blockquote" },
1247 { "p", "body" },
1248 { "p", "caption" },
1249 { "p", "center" },
1250 { "p", "col" },
1251 { "p", "colgroup" },
1252 { "p", "dd" },
1253 { "p", "dir" },
1254 { "p", "div" },
1255 { "p", "dl" },
1256 { "p", "dt" },
1257 { "p", "fieldset" },
1258 { "p", "form" },
1259 { "p", "frameset" },
1260 { "p", "h1" },
1261 { "p", "h2" },
1262 { "p", "h3" },
1263 { "p", "h4" },
1264 { "p", "h5" },
1265 { "p", "h6" },
1266 { "p", "head" },
1267 { "p", "hr" },
1268 { "p", "li" },
1269 { "p", "listing" },
1270 { "p", "menu" },
1271 { "p", "ol" },
1272 { "p", "p" },
1273 { "p", "pre" },
1274 { "p", "table" },
1275 { "p", "tbody" },
1276 { "p", "td" },
1277 { "p", "tfoot" },
1278 { "p", "th" },
1279 { "p", "title" },
1280 { "p", "tr" },
1281 { "p", "ul" },
1282 { "p", "xmp" },
1283 { "pre", "dd" },
1284 { "pre", "dl" },
1285 { "pre", "dt" },
1286 { "pre", "fieldset" },
1287 { "pre", "form" },
1288 { "pre", "li" },
1289 { "pre", "table" },
1290 { "pre", "ul" },
1291 { "s", "p" },
1292 { "script", "noscript" },
1293 { "small", "p" },
1294 { "span", "td" },
1295 { "span", "th" },
1296 { "strike", "p" },
1297 { "style", "body" },
1298 { "style", "frameset" },
1299 { "tbody", "tbody" },
1300 { "tbody", "tfoot" },
1301 { "td", "tbody" },
1302 { "td", "td" },
1303 { "td", "tfoot" },
1304 { "td", "th" },
1305 { "td", "tr" },
1306 { "tfoot", "tbody" },
1307 { "th", "tbody" },
1308 { "th", "td" },
1309 { "th", "tfoot" },
1310 { "th", "th" },
1311 { "th", "tr" },
1312 { "thead", "tbody" },
1313 { "thead", "tfoot" },
1314 { "title", "body" },
1315 { "title", "frameset" },
1316 { "tr", "tbody" },
1317 { "tr", "tfoot" },
1318 { "tr", "tr" },
1319 { "tt", "p" },
1320 { "u", "p" },
1321 { "u", "td" },
1322 { "u", "th" },
1323 { "ul", "address" },
1324 { "ul", "form" },
1325 { "ul", "menu" },
1326 { "ul", "ol" },
1327 { "ul", "pre" },
1328 { "xmp", "dd" },
1329 { "xmp", "dl" },
1330 { "xmp", "dt" },
1331 { "xmp", "fieldset" },
1332 { "xmp", "form" },
1333 { "xmp", "li" },
1334 { "xmp", "table" },
1335 { "xmp", "ul" }
1336};
1337
1338/*
1339 * The list of HTML elements which are supposed not to have
1340 * CDATA content and where a p element will be implied
1341 *
1342 * TODO: extend that list by reading the HTML SGML DTD on
1343 * implied paragraph
1344 */
1345static const char *const htmlNoContentElements[] = {
1346 "html",
1347 "head",
1348 NULL
1349};
1350
1351/*
1352 * The list of HTML attributes which are of content %Script;
1353 * NOTE: when adding ones, check htmlIsScriptAttribute() since
1354 * it assumes the name starts with 'on'
1355 */
1356static const char *const htmlScriptAttributes[] = {
1357 "onclick",
1358 "ondblclick",
1359 "onmousedown",
1360 "onmouseup",
1361 "onmouseover",
1362 "onmousemove",
1363 "onmouseout",
1364 "onkeypress",
1365 "onkeydown",
1366 "onkeyup",
1367 "onload",
1368 "onunload",
1369 "onfocus",
1370 "onblur",
1371 "onsubmit",
1372 "onreset",
1373 "onchange",
1374 "onselect"
1375};
1376
1377/*
1378 * This table is used by the htmlparser to know what to do with
1379 * broken html pages. By assigning different priorities to different
1380 * elements the parser can decide how to handle extra endtags.
1381 * Endtags are only allowed to close elements with lower or equal
1382 * priority.
1383 */
1384
1385typedef struct {
1386 const char *name;
1387 int priority;
1388} elementPriority;
1389
1390static const elementPriority htmlEndPriority[] = {
1391 {"div", 150},
1392 {"td", 160},
1393 {"th", 160},
1394 {"tr", 170},
1395 {"thead", 180},
1396 {"tbody", 180},
1397 {"tfoot", 180},
1398 {"table", 190},
1399 {"head", 200},
1400 {"body", 200},
1401 {"html", 220},
1402 {NULL, 100} /* Default priority */
1403};
1404
1405/************************************************************************
1406 * *
1407 * functions to handle HTML specific data *
1408 * *
1409 ************************************************************************/
1410
1411/**
1412 * htmlInitAutoClose:
1413 *
1414 * This is a no-op now.
1415 */
1416void
1417htmlInitAutoClose(void) {
1418}
1419
1420static int
1421htmlCompareTags(const void *key, const void *member) {
1422 const xmlChar *tag = (const xmlChar *) key;
1423 const htmlElemDesc *desc = (const htmlElemDesc *) member;
1424
1425 return(xmlStrcasecmp(tag, BAD_CAST desc->name));
1426}
1427
1428/**
1429 * htmlTagLookup:
1430 * @tag: The tag name in lowercase
1431 *
1432 * Lookup the HTML tag in the ElementTable
1433 *
1434 * Returns the related htmlElemDescPtr or NULL if not found.
1435 */
1436const htmlElemDesc *
1437htmlTagLookup(const xmlChar *tag) {
1438 if (tag == NULL)
1439 return(NULL);
1440
1441 return((const htmlElemDesc *) bsearch(tag, html40ElementTable,
1442 sizeof(html40ElementTable) / sizeof(htmlElemDesc),
1443 sizeof(htmlElemDesc), htmlCompareTags));
1444}
1445
1446/**
1447 * htmlGetEndPriority:
1448 * @name: The name of the element to look up the priority for.
1449 *
1450 * Return value: The "endtag" priority.
1451 **/
1452static int
1453htmlGetEndPriority (const xmlChar *name) {
1454 int i = 0;
1455
1456 while ((htmlEndPriority[i].name != NULL) &&
1457 (!xmlStrEqual((const xmlChar *)htmlEndPriority[i].name, name)))
1458 i++;
1459
1460 return(htmlEndPriority[i].priority);
1461}
1462
1463
1464static int
1465htmlCompareStartClose(const void *vkey, const void *member) {
1466 const htmlStartCloseEntry *key = (const htmlStartCloseEntry *) vkey;
1467 const htmlStartCloseEntry *entry = (const htmlStartCloseEntry *) member;
1468 int ret;
1469
1470 ret = strcmp(key->oldTag, entry->oldTag);
1471 if (ret == 0)
1472 ret = strcmp(key->newTag, entry->newTag);
1473
1474 return(ret);
1475}
1476
1477/**
1478 * htmlCheckAutoClose:
1479 * @newtag: The new tag name
1480 * @oldtag: The old tag name
1481 *
1482 * Checks whether the new tag is one of the registered valid tags for
1483 * closing old.
1484 *
1485 * Returns 0 if no, 1 if yes.
1486 */
1487static int
1488htmlCheckAutoClose(const xmlChar * newtag, const xmlChar * oldtag)
1489{
1490 htmlStartCloseEntry key;
1491 void *res;
1492
1493 key.oldTag = (const char *) oldtag;
1494 key.newTag = (const char *) newtag;
1495 res = bsearch(&key, htmlStartClose,
1496 sizeof(htmlStartClose) / sizeof(htmlStartCloseEntry),
1497 sizeof(htmlStartCloseEntry), htmlCompareStartClose);
1498 return(res != NULL);
1499}
1500
1501/**
1502 * htmlAutoCloseOnClose:
1503 * @ctxt: an HTML parser context
1504 * @newtag: The new tag name
1505 * @force: force the tag closure
1506 *
1507 * The HTML DTD allows an ending tag to implicitly close other tags.
1508 */
1509static void
1510htmlAutoCloseOnClose(htmlParserCtxtPtr ctxt, const xmlChar * newtag)
1511{
1512 const htmlElemDesc *info;
1513 int i, priority;
1514
1515 priority = htmlGetEndPriority(newtag);
1516
1517 for (i = (ctxt->nameNr - 1); i >= 0; i--) {
1518
1519 if (xmlStrEqual(newtag, ctxt->nameTab[i]))
1520 break;
1521 /*
1522 * A misplaced endtag can only close elements with lower
1523 * or equal priority, so if we find an element with higher
1524 * priority before we find an element with
1525 * matching name, we just ignore this endtag
1526 */
1527 if (htmlGetEndPriority(ctxt->nameTab[i]) > priority)
1528 return;
1529 }
1530 if (i < 0)
1531 return;
1532
1533 while (!xmlStrEqual(newtag, ctxt->name)) {
1534 info = htmlTagLookup(ctxt->name);
1535 if ((info != NULL) && (info->endTag == 3)) {
1536 htmlParseErr(ctxt, XML_ERR_TAG_NAME_MISMATCH,
1537 "Opening and ending tag mismatch: %s and %s\n",
1538 newtag, ctxt->name);
1539 }
1540 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
1541 ctxt->sax->endElement(ctxt->userData, ctxt->name);
1542 htmlnamePop(ctxt);
1543 }
1544}
1545
1546/**
1547 * htmlAutoCloseOnEnd:
1548 * @ctxt: an HTML parser context
1549 *
1550 * Close all remaining tags at the end of the stream
1551 */
1552static void
1553htmlAutoCloseOnEnd(htmlParserCtxtPtr ctxt)
1554{
1555 int i;
1556
1557 if (ctxt->nameNr == 0)
1558 return;
1559 for (i = (ctxt->nameNr - 1); i >= 0; i--) {
1560 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
1561 ctxt->sax->endElement(ctxt->userData, ctxt->name);
1562 htmlnamePop(ctxt);
1563 }
1564}
1565
1566/**
1567 * htmlAutoClose:
1568 * @ctxt: an HTML parser context
1569 * @newtag: The new tag name or NULL
1570 *
1571 * The HTML DTD allows a tag to implicitly close other tags.
1572 * The list is kept in htmlStartClose array. This function is
1573 * called when a new tag has been detected and generates the
1574 * appropriates closes if possible/needed.
1575 * If newtag is NULL this mean we are at the end of the resource
1576 * and we should check
1577 */
1578static void
1579htmlAutoClose(htmlParserCtxtPtr ctxt, const xmlChar * newtag)
1580{
1581 while ((newtag != NULL) && (ctxt->name != NULL) &&
1582 (htmlCheckAutoClose(newtag, ctxt->name))) {
1583 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
1584 ctxt->sax->endElement(ctxt->userData, ctxt->name);
1585 htmlnamePop(ctxt);
1586 }
1587 if (newtag == NULL) {
1588 htmlAutoCloseOnEnd(ctxt);
1589 return;
1590 }
1591 while ((newtag == NULL) && (ctxt->name != NULL) &&
1592 ((xmlStrEqual(ctxt->name, BAD_CAST "head")) ||
1593 (xmlStrEqual(ctxt->name, BAD_CAST "body")) ||
1594 (xmlStrEqual(ctxt->name, BAD_CAST "html")))) {
1595 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
1596 ctxt->sax->endElement(ctxt->userData, ctxt->name);
1597 htmlnamePop(ctxt);
1598 }
1599}
1600
1601/**
1602 * htmlAutoCloseTag:
1603 * @doc: the HTML document
1604 * @name: The tag name
1605 * @elem: the HTML element
1606 *
1607 * The HTML DTD allows a tag to implicitly close other tags.
1608 * The list is kept in htmlStartClose array. This function checks
1609 * if the element or one of it's children would autoclose the
1610 * given tag.
1611 *
1612 * Returns 1 if autoclose, 0 otherwise
1613 */
1614int
1615htmlAutoCloseTag(htmlDocPtr doc, const xmlChar *name, htmlNodePtr elem) {
1616 htmlNodePtr child;
1617
1618 if (elem == NULL) return(1);
1619 if (xmlStrEqual(name, elem->name)) return(0);
1620 if (htmlCheckAutoClose(elem->name, name)) return(1);
1621 child = elem->children;
1622 while (child != NULL) {
1623 if (htmlAutoCloseTag(doc, name, child)) return(1);
1624 child = child->next;
1625 }
1626 return(0);
1627}
1628
1629/**
1630 * htmlIsAutoClosed:
1631 * @doc: the HTML document
1632 * @elem: the HTML element
1633 *
1634 * The HTML DTD allows a tag to implicitly close other tags.
1635 * The list is kept in htmlStartClose array. This function checks
1636 * if a tag is autoclosed by one of it's child
1637 *
1638 * Returns 1 if autoclosed, 0 otherwise
1639 */
1640int
1641htmlIsAutoClosed(htmlDocPtr doc, htmlNodePtr elem) {
1642 htmlNodePtr child;
1643
1644 if (elem == NULL) return(1);
1645 child = elem->children;
1646 while (child != NULL) {
1647 if (htmlAutoCloseTag(doc, elem->name, child)) return(1);
1648 child = child->next;
1649 }
1650 return(0);
1651}
1652
1653/**
1654 * htmlCheckImplied:
1655 * @ctxt: an HTML parser context
1656 * @newtag: The new tag name
1657 *
1658 * The HTML DTD allows a tag to exists only implicitly
1659 * called when a new tag has been detected and generates the
1660 * appropriates implicit tags if missing
1661 */
1662static void
1663htmlCheckImplied(htmlParserCtxtPtr ctxt, const xmlChar *newtag) {
1664 int i;
1665
1666 if (ctxt->options & HTML_PARSE_NOIMPLIED)
1667 return;
1668 if (!htmlOmittedDefaultValue)
1669 return;
1670 if (xmlStrEqual(newtag, BAD_CAST"html"))
1671 return;
1672 if (ctxt->nameNr <= 0) {
1673 htmlnamePush(ctxt, BAD_CAST"html");
1674 if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
1675 ctxt->sax->startElement(ctxt->userData, BAD_CAST"html", NULL);
1676 }
1677 if ((xmlStrEqual(newtag, BAD_CAST"body")) || (xmlStrEqual(newtag, BAD_CAST"head")))
1678 return;
1679 if ((ctxt->nameNr <= 1) &&
1680 ((xmlStrEqual(newtag, BAD_CAST"script")) ||
1681 (xmlStrEqual(newtag, BAD_CAST"style")) ||
1682 (xmlStrEqual(newtag, BAD_CAST"meta")) ||
1683 (xmlStrEqual(newtag, BAD_CAST"link")) ||
1684 (xmlStrEqual(newtag, BAD_CAST"title")) ||
1685 (xmlStrEqual(newtag, BAD_CAST"base")))) {
1686 if (ctxt->html >= 3) {
1687 /* we already saw or generated an <head> before */
1688 return;
1689 }
1690 /*
1691 * dropped OBJECT ... i you put it first BODY will be
1692 * assumed !
1693 */
1694 htmlnamePush(ctxt, BAD_CAST"head");
1695 if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
1696 ctxt->sax->startElement(ctxt->userData, BAD_CAST"head", NULL);
1697 } else if ((!xmlStrEqual(newtag, BAD_CAST"noframes")) &&
1698 (!xmlStrEqual(newtag, BAD_CAST"frame")) &&
1699 (!xmlStrEqual(newtag, BAD_CAST"frameset"))) {
1700 if (ctxt->html >= 10) {
1701 /* we already saw or generated a <body> before */
1702 return;
1703 }
1704 for (i = 0;i < ctxt->nameNr;i++) {
1705 if (xmlStrEqual(ctxt->nameTab[i], BAD_CAST"body")) {
1706 return;
1707 }
1708 if (xmlStrEqual(ctxt->nameTab[i], BAD_CAST"head")) {
1709 return;
1710 }
1711 }
1712
1713 htmlnamePush(ctxt, BAD_CAST"body");
1714 if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
1715 ctxt->sax->startElement(ctxt->userData, BAD_CAST"body", NULL);
1716 }
1717}
1718
1719/**
1720 * htmlCheckParagraph
1721 * @ctxt: an HTML parser context
1722 *
1723 * Check whether a p element need to be implied before inserting
1724 * characters in the current element.
1725 *
1726 * Returns 1 if a paragraph has been inserted, 0 if not and -1
1727 * in case of error.
1728 */
1729
1730static int
1731htmlCheckParagraph(htmlParserCtxtPtr ctxt) {
1732 const xmlChar *tag;
1733 int i;
1734
1735 if (ctxt == NULL)
1736 return(-1);
1737 tag = ctxt->name;
1738 if (tag == NULL) {
1739 htmlAutoClose(ctxt, BAD_CAST"p");
1740 htmlCheckImplied(ctxt, BAD_CAST"p");
1741 htmlnamePush(ctxt, BAD_CAST"p");
1742 if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
1743 ctxt->sax->startElement(ctxt->userData, BAD_CAST"p", NULL);
1744 return(1);
1745 }
1746 if (!htmlOmittedDefaultValue)
1747 return(0);
1748 for (i = 0; htmlNoContentElements[i] != NULL; i++) {
1749 if (xmlStrEqual(tag, BAD_CAST htmlNoContentElements[i])) {
1750 htmlAutoClose(ctxt, BAD_CAST"p");
1751 htmlCheckImplied(ctxt, BAD_CAST"p");
1752 htmlnamePush(ctxt, BAD_CAST"p");
1753 if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
1754 ctxt->sax->startElement(ctxt->userData, BAD_CAST"p", NULL);
1755 return(1);
1756 }
1757 }
1758 return(0);
1759}
1760
1761/**
1762 * htmlIsScriptAttribute:
1763 * @name: an attribute name
1764 *
1765 * Check if an attribute is of content type Script
1766 *
1767 * Returns 1 is the attribute is a script 0 otherwise
1768 */
1769int
1770htmlIsScriptAttribute(const xmlChar *name) {
1771 unsigned int i;
1772
1773 if (name == NULL)
1774 return(0);
1775 /*
1776 * all script attributes start with 'on'
1777 */
1778 if ((name[0] != 'o') || (name[1] != 'n'))
1779 return(0);
1780 for (i = 0;
1781 i < sizeof(htmlScriptAttributes)/sizeof(htmlScriptAttributes[0]);
1782 i++) {
1783 if (xmlStrEqual(name, (const xmlChar *) htmlScriptAttributes[i]))
1784 return(1);
1785 }
1786 return(0);
1787}
1788
1789/************************************************************************
1790 * *
1791 * The list of HTML predefined entities *
1792 * *
1793 ************************************************************************/
1794
1795
1796static const htmlEntityDesc html40EntitiesTable[] = {
1797/*
1798 * the 4 absolute ones, plus apostrophe.
1799 */
1800{ 34, "quot", "quotation mark = APL quote, U+0022 ISOnum" },
1801{ 38, "amp", "ampersand, U+0026 ISOnum" },
1802{ 39, "apos", "single quote" },
1803{ 60, "lt", "less-than sign, U+003C ISOnum" },
1804{ 62, "gt", "greater-than sign, U+003E ISOnum" },
1805
1806/*
1807 * A bunch still in the 128-255 range
1808 * Replacing them depend really on the charset used.
1809 */
1810{ 160, "nbsp", "no-break space = non-breaking space, U+00A0 ISOnum" },
1811{ 161, "iexcl","inverted exclamation mark, U+00A1 ISOnum" },
1812{ 162, "cent", "cent sign, U+00A2 ISOnum" },
1813{ 163, "pound","pound sign, U+00A3 ISOnum" },
1814{ 164, "curren","currency sign, U+00A4 ISOnum" },
1815{ 165, "yen", "yen sign = yuan sign, U+00A5 ISOnum" },
1816{ 166, "brvbar","broken bar = broken vertical bar, U+00A6 ISOnum" },
1817{ 167, "sect", "section sign, U+00A7 ISOnum" },
1818{ 168, "uml", "diaeresis = spacing diaeresis, U+00A8 ISOdia" },
1819{ 169, "copy", "copyright sign, U+00A9 ISOnum" },
1820{ 170, "ordf", "feminine ordinal indicator, U+00AA ISOnum" },
1821{ 171, "laquo","left-pointing double angle quotation mark = left pointing guillemet, U+00AB ISOnum" },
1822{ 172, "not", "not sign, U+00AC ISOnum" },
1823{ 173, "shy", "soft hyphen = discretionary hyphen, U+00AD ISOnum" },
1824{ 174, "reg", "registered sign = registered trade mark sign, U+00AE ISOnum" },
1825{ 175, "macr", "macron = spacing macron = overline = APL overbar, U+00AF ISOdia" },
1826{ 176, "deg", "degree sign, U+00B0 ISOnum" },
1827{ 177, "plusmn","plus-minus sign = plus-or-minus sign, U+00B1 ISOnum" },
1828{ 178, "sup2", "superscript two = superscript digit two = squared, U+00B2 ISOnum" },
1829{ 179, "sup3", "superscript three = superscript digit three = cubed, U+00B3 ISOnum" },
1830{ 180, "acute","acute accent = spacing acute, U+00B4 ISOdia" },
1831{ 181, "micro","micro sign, U+00B5 ISOnum" },
1832{ 182, "para", "pilcrow sign = paragraph sign, U+00B6 ISOnum" },
1833{ 183, "middot","middle dot = Georgian comma Greek middle dot, U+00B7 ISOnum" },
1834{ 184, "cedil","cedilla = spacing cedilla, U+00B8 ISOdia" },
1835{ 185, "sup1", "superscript one = superscript digit one, U+00B9 ISOnum" },
1836{ 186, "ordm", "masculine ordinal indicator, U+00BA ISOnum" },
1837{ 187, "raquo","right-pointing double angle quotation mark right pointing guillemet, U+00BB ISOnum" },
1838{ 188, "frac14","vulgar fraction one quarter = fraction one quarter, U+00BC ISOnum" },
1839{ 189, "frac12","vulgar fraction one half = fraction one half, U+00BD ISOnum" },
1840{ 190, "frac34","vulgar fraction three quarters = fraction three quarters, U+00BE ISOnum" },
1841{ 191, "iquest","inverted question mark = turned question mark, U+00BF ISOnum" },
1842{ 192, "Agrave","latin capital letter A with grave = latin capital letter A grave, U+00C0 ISOlat1" },
1843{ 193, "Aacute","latin capital letter A with acute, U+00C1 ISOlat1" },
1844{ 194, "Acirc","latin capital letter A with circumflex, U+00C2 ISOlat1" },
1845{ 195, "Atilde","latin capital letter A with tilde, U+00C3 ISOlat1" },
1846{ 196, "Auml", "latin capital letter A with diaeresis, U+00C4 ISOlat1" },
1847{ 197, "Aring","latin capital letter A with ring above = latin capital letter A ring, U+00C5 ISOlat1" },
1848{ 198, "AElig","latin capital letter AE = latin capital ligature AE, U+00C6 ISOlat1" },
1849{ 199, "Ccedil","latin capital letter C with cedilla, U+00C7 ISOlat1" },
1850{ 200, "Egrave","latin capital letter E with grave, U+00C8 ISOlat1" },
1851{ 201, "Eacute","latin capital letter E with acute, U+00C9 ISOlat1" },
1852{ 202, "Ecirc","latin capital letter E with circumflex, U+00CA ISOlat1" },
1853{ 203, "Euml", "latin capital letter E with diaeresis, U+00CB ISOlat1" },
1854{ 204, "Igrave","latin capital letter I with grave, U+00CC ISOlat1" },
1855{ 205, "Iacute","latin capital letter I with acute, U+00CD ISOlat1" },
1856{ 206, "Icirc","latin capital letter I with circumflex, U+00CE ISOlat1" },
1857{ 207, "Iuml", "latin capital letter I with diaeresis, U+00CF ISOlat1" },
1858{ 208, "ETH", "latin capital letter ETH, U+00D0 ISOlat1" },
1859{ 209, "Ntilde","latin capital letter N with tilde, U+00D1 ISOlat1" },
1860{ 210, "Ograve","latin capital letter O with grave, U+00D2 ISOlat1" },
1861{ 211, "Oacute","latin capital letter O with acute, U+00D3 ISOlat1" },
1862{ 212, "Ocirc","latin capital letter O with circumflex, U+00D4 ISOlat1" },
1863{ 213, "Otilde","latin capital letter O with tilde, U+00D5 ISOlat1" },
1864{ 214, "Ouml", "latin capital letter O with diaeresis, U+00D6 ISOlat1" },
1865{ 215, "times","multiplication sign, U+00D7 ISOnum" },
1866{ 216, "Oslash","latin capital letter O with stroke latin capital letter O slash, U+00D8 ISOlat1" },
1867{ 217, "Ugrave","latin capital letter U with grave, U+00D9 ISOlat1" },
1868{ 218, "Uacute","latin capital letter U with acute, U+00DA ISOlat1" },
1869{ 219, "Ucirc","latin capital letter U with circumflex, U+00DB ISOlat1" },
1870{ 220, "Uuml", "latin capital letter U with diaeresis, U+00DC ISOlat1" },
1871{ 221, "Yacute","latin capital letter Y with acute, U+00DD ISOlat1" },
1872{ 222, "THORN","latin capital letter THORN, U+00DE ISOlat1" },
1873{ 223, "szlig","latin small letter sharp s = ess-zed, U+00DF ISOlat1" },
1874{ 224, "agrave","latin small letter a with grave = latin small letter a grave, U+00E0 ISOlat1" },
1875{ 225, "aacute","latin small letter a with acute, U+00E1 ISOlat1" },
1876{ 226, "acirc","latin small letter a with circumflex, U+00E2 ISOlat1" },
1877{ 227, "atilde","latin small letter a with tilde, U+00E3 ISOlat1" },
1878{ 228, "auml", "latin small letter a with diaeresis, U+00E4 ISOlat1" },
1879{ 229, "aring","latin small letter a with ring above = latin small letter a ring, U+00E5 ISOlat1" },
1880{ 230, "aelig","latin small letter ae = latin small ligature ae, U+00E6 ISOlat1" },
1881{ 231, "ccedil","latin small letter c with cedilla, U+00E7 ISOlat1" },
1882{ 232, "egrave","latin small letter e with grave, U+00E8 ISOlat1" },
1883{ 233, "eacute","latin small letter e with acute, U+00E9 ISOlat1" },
1884{ 234, "ecirc","latin small letter e with circumflex, U+00EA ISOlat1" },
1885{ 235, "euml", "latin small letter e with diaeresis, U+00EB ISOlat1" },
1886{ 236, "igrave","latin small letter i with grave, U+00EC ISOlat1" },
1887{ 237, "iacute","latin small letter i with acute, U+00ED ISOlat1" },
1888{ 238, "icirc","latin small letter i with circumflex, U+00EE ISOlat1" },
1889{ 239, "iuml", "latin small letter i with diaeresis, U+00EF ISOlat1" },
1890{ 240, "eth", "latin small letter eth, U+00F0 ISOlat1" },
1891{ 241, "ntilde","latin small letter n with tilde, U+00F1 ISOlat1" },
1892{ 242, "ograve","latin small letter o with grave, U+00F2 ISOlat1" },
1893{ 243, "oacute","latin small letter o with acute, U+00F3 ISOlat1" },
1894{ 244, "ocirc","latin small letter o with circumflex, U+00F4 ISOlat1" },
1895{ 245, "otilde","latin small letter o with tilde, U+00F5 ISOlat1" },
1896{ 246, "ouml", "latin small letter o with diaeresis, U+00F6 ISOlat1" },
1897{ 247, "divide","division sign, U+00F7 ISOnum" },
1898{ 248, "oslash","latin small letter o with stroke, = latin small letter o slash, U+00F8 ISOlat1" },
1899{ 249, "ugrave","latin small letter u with grave, U+00F9 ISOlat1" },
1900{ 250, "uacute","latin small letter u with acute, U+00FA ISOlat1" },
1901{ 251, "ucirc","latin small letter u with circumflex, U+00FB ISOlat1" },
1902{ 252, "uuml", "latin small letter u with diaeresis, U+00FC ISOlat1" },
1903{ 253, "yacute","latin small letter y with acute, U+00FD ISOlat1" },
1904{ 254, "thorn","latin small letter thorn with, U+00FE ISOlat1" },
1905{ 255, "yuml", "latin small letter y with diaeresis, U+00FF ISOlat1" },
1906
1907{ 338, "OElig","latin capital ligature OE, U+0152 ISOlat2" },
1908{ 339, "oelig","latin small ligature oe, U+0153 ISOlat2" },
1909{ 352, "Scaron","latin capital letter S with caron, U+0160 ISOlat2" },
1910{ 353, "scaron","latin small letter s with caron, U+0161 ISOlat2" },
1911{ 376, "Yuml", "latin capital letter Y with diaeresis, U+0178 ISOlat2" },
1912
1913/*
1914 * Anything below should really be kept as entities references
1915 */
1916{ 402, "fnof", "latin small f with hook = function = florin, U+0192 ISOtech" },
1917
1918{ 710, "circ", "modifier letter circumflex accent, U+02C6 ISOpub" },
1919{ 732, "tilde","small tilde, U+02DC ISOdia" },
1920
1921{ 913, "Alpha","greek capital letter alpha, U+0391" },
1922{ 914, "Beta", "greek capital letter beta, U+0392" },
1923{ 915, "Gamma","greek capital letter gamma, U+0393 ISOgrk3" },
1924{ 916, "Delta","greek capital letter delta, U+0394 ISOgrk3" },
1925{ 917, "Epsilon","greek capital letter epsilon, U+0395" },
1926{ 918, "Zeta", "greek capital letter zeta, U+0396" },
1927{ 919, "Eta", "greek capital letter eta, U+0397" },
1928{ 920, "Theta","greek capital letter theta, U+0398 ISOgrk3" },
1929{ 921, "Iota", "greek capital letter iota, U+0399" },
1930{ 922, "Kappa","greek capital letter kappa, U+039A" },
1931{ 923, "Lambda", "greek capital letter lambda, U+039B ISOgrk3" },
1932{ 924, "Mu", "greek capital letter mu, U+039C" },
1933{ 925, "Nu", "greek capital letter nu, U+039D" },
1934{ 926, "Xi", "greek capital letter xi, U+039E ISOgrk3" },
1935{ 927, "Omicron","greek capital letter omicron, U+039F" },
1936{ 928, "Pi", "greek capital letter pi, U+03A0 ISOgrk3" },
1937{ 929, "Rho", "greek capital letter rho, U+03A1" },
1938{ 931, "Sigma","greek capital letter sigma, U+03A3 ISOgrk3" },
1939{ 932, "Tau", "greek capital letter tau, U+03A4" },
1940{ 933, "Upsilon","greek capital letter upsilon, U+03A5 ISOgrk3" },
1941{ 934, "Phi", "greek capital letter phi, U+03A6 ISOgrk3" },
1942{ 935, "Chi", "greek capital letter chi, U+03A7" },
1943{ 936, "Psi", "greek capital letter psi, U+03A8 ISOgrk3" },
1944{ 937, "Omega","greek capital letter omega, U+03A9 ISOgrk3" },
1945
1946{ 945, "alpha","greek small letter alpha, U+03B1 ISOgrk3" },
1947{ 946, "beta", "greek small letter beta, U+03B2 ISOgrk3" },
1948{ 947, "gamma","greek small letter gamma, U+03B3 ISOgrk3" },
1949{ 948, "delta","greek small letter delta, U+03B4 ISOgrk3" },
1950{ 949, "epsilon","greek small letter epsilon, U+03B5 ISOgrk3" },
1951{ 950, "zeta", "greek small letter zeta, U+03B6 ISOgrk3" },
1952{ 951, "eta", "greek small letter eta, U+03B7 ISOgrk3" },
1953{ 952, "theta","greek small letter theta, U+03B8 ISOgrk3" },
1954{ 953, "iota", "greek small letter iota, U+03B9 ISOgrk3" },
1955{ 954, "kappa","greek small letter kappa, U+03BA ISOgrk3" },
1956{ 955, "lambda","greek small letter lambda, U+03BB ISOgrk3" },
1957{ 956, "mu", "greek small letter mu, U+03BC ISOgrk3" },
1958{ 957, "nu", "greek small letter nu, U+03BD ISOgrk3" },
1959{ 958, "xi", "greek small letter xi, U+03BE ISOgrk3" },
1960{ 959, "omicron","greek small letter omicron, U+03BF NEW" },
1961{ 960, "pi", "greek small letter pi, U+03C0 ISOgrk3" },
1962{ 961, "rho", "greek small letter rho, U+03C1 ISOgrk3" },
1963{ 962, "sigmaf","greek small letter final sigma, U+03C2 ISOgrk3" },
1964{ 963, "sigma","greek small letter sigma, U+03C3 ISOgrk3" },
1965{ 964, "tau", "greek small letter tau, U+03C4 ISOgrk3" },
1966{ 965, "upsilon","greek small letter upsilon, U+03C5 ISOgrk3" },
1967{ 966, "phi", "greek small letter phi, U+03C6 ISOgrk3" },
1968{ 967, "chi", "greek small letter chi, U+03C7 ISOgrk3" },
1969{ 968, "psi", "greek small letter psi, U+03C8 ISOgrk3" },
1970{ 969, "omega","greek small letter omega, U+03C9 ISOgrk3" },
1971{ 977, "thetasym","greek small letter theta symbol, U+03D1 NEW" },
1972{ 978, "upsih","greek upsilon with hook symbol, U+03D2 NEW" },
1973{ 982, "piv", "greek pi symbol, U+03D6 ISOgrk3" },
1974
1975{ 8194, "ensp", "en space, U+2002 ISOpub" },
1976{ 8195, "emsp", "em space, U+2003 ISOpub" },
1977{ 8201, "thinsp","thin space, U+2009 ISOpub" },
1978{ 8204, "zwnj", "zero width non-joiner, U+200C NEW RFC 2070" },
1979{ 8205, "zwj", "zero width joiner, U+200D NEW RFC 2070" },
1980{ 8206, "lrm", "left-to-right mark, U+200E NEW RFC 2070" },
1981{ 8207, "rlm", "right-to-left mark, U+200F NEW RFC 2070" },
1982{ 8211, "ndash","en dash, U+2013 ISOpub" },
1983{ 8212, "mdash","em dash, U+2014 ISOpub" },
1984{ 8216, "lsquo","left single quotation mark, U+2018 ISOnum" },
1985{ 8217, "rsquo","right single quotation mark, U+2019 ISOnum" },
1986{ 8218, "sbquo","single low-9 quotation mark, U+201A NEW" },
1987{ 8220, "ldquo","left double quotation mark, U+201C ISOnum" },
1988{ 8221, "rdquo","right double quotation mark, U+201D ISOnum" },
1989{ 8222, "bdquo","double low-9 quotation mark, U+201E NEW" },
1990{ 8224, "dagger","dagger, U+2020 ISOpub" },
1991{ 8225, "Dagger","double dagger, U+2021 ISOpub" },
1992
1993{ 8226, "bull", "bullet = black small circle, U+2022 ISOpub" },
1994{ 8230, "hellip","horizontal ellipsis = three dot leader, U+2026 ISOpub" },
1995
1996{ 8240, "permil","per mille sign, U+2030 ISOtech" },
1997
1998{ 8242, "prime","prime = minutes = feet, U+2032 ISOtech" },
1999{ 8243, "Prime","double prime = seconds = inches, U+2033 ISOtech" },
2000
2001{ 8249, "lsaquo","single left-pointing angle quotation mark, U+2039 ISO proposed" },
2002{ 8250, "rsaquo","single right-pointing angle quotation mark, U+203A ISO proposed" },
2003
2004{ 8254, "oline","overline = spacing overscore, U+203E NEW" },
2005{ 8260, "frasl","fraction slash, U+2044 NEW" },
2006
2007{ 8364, "euro", "euro sign, U+20AC NEW" },
2008
2009{ 8465, "image","blackletter capital I = imaginary part, U+2111 ISOamso" },
2010{ 8472, "weierp","script capital P = power set = Weierstrass p, U+2118 ISOamso" },
2011{ 8476, "real", "blackletter capital R = real part symbol, U+211C ISOamso" },
2012{ 8482, "trade","trade mark sign, U+2122 ISOnum" },
2013{ 8501, "alefsym","alef symbol = first transfinite cardinal, U+2135 NEW" },
2014{ 8592, "larr", "leftwards arrow, U+2190 ISOnum" },
2015{ 8593, "uarr", "upwards arrow, U+2191 ISOnum" },
2016{ 8594, "rarr", "rightwards arrow, U+2192 ISOnum" },
2017{ 8595, "darr", "downwards arrow, U+2193 ISOnum" },
2018{ 8596, "harr", "left right arrow, U+2194 ISOamsa" },
2019{ 8629, "crarr","downwards arrow with corner leftwards = carriage return, U+21B5 NEW" },
2020{ 8656, "lArr", "leftwards double arrow, U+21D0 ISOtech" },
2021{ 8657, "uArr", "upwards double arrow, U+21D1 ISOamsa" },
2022{ 8658, "rArr", "rightwards double arrow, U+21D2 ISOtech" },
2023{ 8659, "dArr", "downwards double arrow, U+21D3 ISOamsa" },
2024{ 8660, "hArr", "left right double arrow, U+21D4 ISOamsa" },
2025
2026{ 8704, "forall","for all, U+2200 ISOtech" },
2027{ 8706, "part", "partial differential, U+2202 ISOtech" },
2028{ 8707, "exist","there exists, U+2203 ISOtech" },
2029{ 8709, "empty","empty set = null set = diameter, U+2205 ISOamso" },
2030{ 8711, "nabla","nabla = backward difference, U+2207 ISOtech" },
2031{ 8712, "isin", "element of, U+2208 ISOtech" },
2032{ 8713, "notin","not an element of, U+2209 ISOtech" },
2033{ 8715, "ni", "contains as member, U+220B ISOtech" },
2034{ 8719, "prod", "n-ary product = product sign, U+220F ISOamsb" },
2035{ 8721, "sum", "n-ary summation, U+2211 ISOamsb" },
2036{ 8722, "minus","minus sign, U+2212 ISOtech" },
2037{ 8727, "lowast","asterisk operator, U+2217 ISOtech" },
2038{ 8730, "radic","square root = radical sign, U+221A ISOtech" },
2039{ 8733, "prop", "proportional to, U+221D ISOtech" },
2040{ 8734, "infin","infinity, U+221E ISOtech" },
2041{ 8736, "ang", "angle, U+2220 ISOamso" },
2042{ 8743, "and", "logical and = wedge, U+2227 ISOtech" },
2043{ 8744, "or", "logical or = vee, U+2228 ISOtech" },
2044{ 8745, "cap", "intersection = cap, U+2229 ISOtech" },
2045{ 8746, "cup", "union = cup, U+222A ISOtech" },
2046{ 8747, "int", "integral, U+222B ISOtech" },
2047{ 8756, "there4","therefore, U+2234 ISOtech" },
2048{ 8764, "sim", "tilde operator = varies with = similar to, U+223C ISOtech" },
2049{ 8773, "cong", "approximately equal to, U+2245 ISOtech" },
2050{ 8776, "asymp","almost equal to = asymptotic to, U+2248 ISOamsr" },
2051{ 8800, "ne", "not equal to, U+2260 ISOtech" },
2052{ 8801, "equiv","identical to, U+2261 ISOtech" },
2053{ 8804, "le", "less-than or equal to, U+2264 ISOtech" },
2054{ 8805, "ge", "greater-than or equal to, U+2265 ISOtech" },
2055{ 8834, "sub", "subset of, U+2282 ISOtech" },
2056{ 8835, "sup", "superset of, U+2283 ISOtech" },
2057{ 8836, "nsub", "not a subset of, U+2284 ISOamsn" },
2058{ 8838, "sube", "subset of or equal to, U+2286 ISOtech" },
2059{ 8839, "supe", "superset of or equal to, U+2287 ISOtech" },
2060{ 8853, "oplus","circled plus = direct sum, U+2295 ISOamsb" },
2061{ 8855, "otimes","circled times = vector product, U+2297 ISOamsb" },
2062{ 8869, "perp", "up tack = orthogonal to = perpendicular, U+22A5 ISOtech" },
2063{ 8901, "sdot", "dot operator, U+22C5 ISOamsb" },
2064{ 8968, "lceil","left ceiling = apl upstile, U+2308 ISOamsc" },
2065{ 8969, "rceil","right ceiling, U+2309 ISOamsc" },
2066{ 8970, "lfloor","left floor = apl downstile, U+230A ISOamsc" },
2067{ 8971, "rfloor","right floor, U+230B ISOamsc" },
2068{ 9001, "lang", "left-pointing angle bracket = bra, U+2329 ISOtech" },
2069{ 9002, "rang", "right-pointing angle bracket = ket, U+232A ISOtech" },
2070{ 9674, "loz", "lozenge, U+25CA ISOpub" },
2071
2072{ 9824, "spades","black spade suit, U+2660 ISOpub" },
2073{ 9827, "clubs","black club suit = shamrock, U+2663 ISOpub" },
2074{ 9829, "hearts","black heart suit = valentine, U+2665 ISOpub" },
2075{ 9830, "diams","black diamond suit, U+2666 ISOpub" },
2076
2077};
2078
2079/************************************************************************
2080 * *
2081 * Commodity functions to handle entities *
2082 * *
2083 ************************************************************************/
2084
2085/*
2086 * Macro used to grow the current buffer.
2087 */
2088#define growBuffer(buffer) { \
2089 xmlChar *tmp; \
2090 buffer##_size *= 2; \
2091 tmp = (xmlChar *) xmlRealloc(buffer, buffer##_size * sizeof(xmlChar)); \
2092 if (tmp == NULL) { \
2093 htmlErrMemory(ctxt, "growing buffer\n"); \
2094 xmlFree(buffer); \
2095 return(NULL); \
2096 } \
2097 buffer = tmp; \
2098}
2099
2100/**
2101 * htmlEntityLookup:
2102 * @name: the entity name
2103 *
2104 * Lookup the given entity in EntitiesTable
2105 *
2106 * TODO: the linear scan is really ugly, an hash table is really needed.
2107 *
2108 * Returns the associated htmlEntityDescPtr if found, NULL otherwise.
2109 */
2110const htmlEntityDesc *
2111htmlEntityLookup(const xmlChar *name) {
2112 unsigned int i;
2113
2114 for (i = 0;i < (sizeof(html40EntitiesTable)/
2115 sizeof(html40EntitiesTable[0]));i++) {
2116 if (xmlStrEqual(name, BAD_CAST html40EntitiesTable[i].name)) {
2117 return((htmlEntityDescPtr) &html40EntitiesTable[i]);
2118 }
2119 }
2120 return(NULL);
2121}
2122
2123/**
2124 * htmlEntityValueLookup:
2125 * @value: the entity's unicode value
2126 *
2127 * Lookup the given entity in EntitiesTable
2128 *
2129 * TODO: the linear scan is really ugly, an hash table is really needed.
2130 *
2131 * Returns the associated htmlEntityDescPtr if found, NULL otherwise.
2132 */
2133const htmlEntityDesc *
2134htmlEntityValueLookup(unsigned int value) {
2135 unsigned int i;
2136
2137 for (i = 0;i < (sizeof(html40EntitiesTable)/
2138 sizeof(html40EntitiesTable[0]));i++) {
2139 if (html40EntitiesTable[i].value >= value) {
2140 if (html40EntitiesTable[i].value > value)
2141 break;
2142 return((htmlEntityDescPtr) &html40EntitiesTable[i]);
2143 }
2144 }
2145 return(NULL);
2146}
2147
2148/**
2149 * UTF8ToHtml:
2150 * @out: a pointer to an array of bytes to store the result
2151 * @outlen: the length of @out
2152 * @in: a pointer to an array of UTF-8 chars
2153 * @inlen: the length of @in
2154 *
2155 * Take a block of UTF-8 chars in and try to convert it to an ASCII
2156 * plus HTML entities block of chars out.
2157 *
2158 * Returns 0 if success, -2 if the transcoding fails, or -1 otherwise
2159 * The value of @inlen after return is the number of octets consumed
2160 * as the return value is positive, else unpredictable.
2161 * The value of @outlen after return is the number of octets consumed.
2162 */
2163int
2164UTF8ToHtml(unsigned char* out, int *outlen,
2165 const unsigned char* in, int *inlen) {
2166 const unsigned char* processed = in;
2167 const unsigned char* outend;
2168 const unsigned char* outstart = out;
2169 const unsigned char* instart = in;
2170 const unsigned char* inend;
2171 unsigned int c, d;
2172 int trailing;
2173
2174 if ((out == NULL) || (outlen == NULL) || (inlen == NULL)) return(-1);
2175 if (in == NULL) {
2176 /*
2177 * initialization nothing to do
2178 */
2179 *outlen = 0;
2180 *inlen = 0;
2181 return(0);
2182 }
2183 inend = in + (*inlen);
2184 outend = out + (*outlen);
2185 while (in < inend) {
2186 d = *in++;
2187 if (d < 0x80) { c= d; trailing= 0; }
2188 else if (d < 0xC0) {
2189 /* trailing byte in leading position */
2190 *outlen = out - outstart;
2191 *inlen = processed - instart;
2192 return(-2);
2193 } else if (d < 0xE0) { c= d & 0x1F; trailing= 1; }
2194 else if (d < 0xF0) { c= d & 0x0F; trailing= 2; }
2195 else if (d < 0xF8) { c= d & 0x07; trailing= 3; }
2196 else {
2197 /* no chance for this in Ascii */
2198 *outlen = out - outstart;
2199 *inlen = processed - instart;
2200 return(-2);
2201 }
2202
2203 if (inend - in < trailing) {
2204 break;
2205 }
2206
2207 for ( ; trailing; trailing--) {
2208 if ((in >= inend) || (((d= *in++) & 0xC0) != 0x80))
2209 break;
2210 c <<= 6;
2211 c |= d & 0x3F;
2212 }
2213
2214 /* assertion: c is a single UTF-4 value */
2215 if (c < 0x80) {
2216 if (out + 1 >= outend)
2217 break;
2218 *out++ = c;
2219 } else {
2220 int len;
2221 const htmlEntityDesc * ent;
2222 const char *cp;
2223 char nbuf[16];
2224
2225 /*
2226 * Try to lookup a predefined HTML entity for it
2227 */
2228
2229 ent = htmlEntityValueLookup(c);
2230 if (ent == NULL) {
2231 snprintf(nbuf, sizeof(nbuf), "#%u", c);
2232 cp = nbuf;
2233 }
2234 else
2235 cp = ent->name;
2236 len = strlen(cp);
2237 if (out + 2 + len >= outend)
2238 break;
2239 *out++ = '&';
2240 memcpy(out, cp, len);
2241 out += len;
2242 *out++ = ';';
2243 }
2244 processed = in;
2245 }
2246 *outlen = out - outstart;
2247 *inlen = processed - instart;
2248 return(0);
2249}
2250
2251/**
2252 * htmlEncodeEntities:
2253 * @out: a pointer to an array of bytes to store the result
2254 * @outlen: the length of @out
2255 * @in: a pointer to an array of UTF-8 chars
2256 * @inlen: the length of @in
2257 * @quoteChar: the quote character to escape (' or ") or zero.
2258 *
2259 * Take a block of UTF-8 chars in and try to convert it to an ASCII
2260 * plus HTML entities block of chars out.
2261 *
2262 * Returns 0 if success, -2 if the transcoding fails, or -1 otherwise
2263 * The value of @inlen after return is the number of octets consumed
2264 * as the return value is positive, else unpredictable.
2265 * The value of @outlen after return is the number of octets consumed.
2266 */
2267int
2268htmlEncodeEntities(unsigned char* out, int *outlen,
2269 const unsigned char* in, int *inlen, int quoteChar) {
2270 const unsigned char* processed = in;
2271 const unsigned char* outend;
2272 const unsigned char* outstart = out;
2273 const unsigned char* instart = in;
2274 const unsigned char* inend;
2275 unsigned int c, d;
2276 int trailing;
2277
2278 if ((out == NULL) || (outlen == NULL) || (inlen == NULL) || (in == NULL))
2279 return(-1);
2280 outend = out + (*outlen);
2281 inend = in + (*inlen);
2282 while (in < inend) {
2283 d = *in++;
2284 if (d < 0x80) { c= d; trailing= 0; }
2285 else if (d < 0xC0) {
2286 /* trailing byte in leading position */
2287 *outlen = out - outstart;
2288 *inlen = processed - instart;
2289 return(-2);
2290 } else if (d < 0xE0) { c= d & 0x1F; trailing= 1; }
2291 else if (d < 0xF0) { c= d & 0x0F; trailing= 2; }
2292 else if (d < 0xF8) { c= d & 0x07; trailing= 3; }
2293 else {
2294 /* no chance for this in Ascii */
2295 *outlen = out - outstart;
2296 *inlen = processed - instart;
2297 return(-2);
2298 }
2299
2300 if (inend - in < trailing)
2301 break;
2302
2303 while (trailing--) {
2304 if (((d= *in++) & 0xC0) != 0x80) {
2305 *outlen = out - outstart;
2306 *inlen = processed - instart;
2307 return(-2);
2308 }
2309 c <<= 6;
2310 c |= d & 0x3F;
2311 }
2312
2313 /* assertion: c is a single UTF-4 value */
2314 if ((c < 0x80) && (c != (unsigned int) quoteChar) &&
2315 (c != '&') && (c != '<') && (c != '>')) {
2316 if (out >= outend)
2317 break;
2318 *out++ = c;
2319 } else {
2320 const htmlEntityDesc * ent;
2321 const char *cp;
2322 char nbuf[16];
2323 int len;
2324
2325 /*
2326 * Try to lookup a predefined HTML entity for it
2327 */
2328 ent = htmlEntityValueLookup(c);
2329 if (ent == NULL) {
2330 snprintf(nbuf, sizeof(nbuf), "#%u", c);
2331 cp = nbuf;
2332 }
2333 else
2334 cp = ent->name;
2335 len = strlen(cp);
2336 if (out + 2 + len > outend)
2337 break;
2338 *out++ = '&';
2339 memcpy(out, cp, len);
2340 out += len;
2341 *out++ = ';';
2342 }
2343 processed = in;
2344 }
2345 *outlen = out - outstart;
2346 *inlen = processed - instart;
2347 return(0);
2348}
2349
2350/************************************************************************
2351 * *
2352 * Commodity functions to handle streams *
2353 * *
2354 ************************************************************************/
2355
2356#ifdef LIBXML_PUSH_ENABLED
2357/**
2358 * htmlNewInputStream:
2359 * @ctxt: an HTML parser context
2360 *
2361 * Create a new input stream structure
2362 * Returns the new input stream or NULL
2363 */
2364static htmlParserInputPtr
2365htmlNewInputStream(htmlParserCtxtPtr ctxt) {
2366 htmlParserInputPtr input;
2367
2368 input = (xmlParserInputPtr) xmlMalloc(sizeof(htmlParserInput));
2369 if (input == NULL) {
2370 htmlErrMemory(ctxt, "couldn't allocate a new input stream\n");
2371 return(NULL);
2372 }
2373 memset(input, 0, sizeof(htmlParserInput));
2374 input->filename = NULL;
2375 input->directory = NULL;
2376 input->base = NULL;
2377 input->cur = NULL;
2378 input->buf = NULL;
2379 input->line = 1;
2380 input->col = 1;
2381 input->buf = NULL;
2382 input->free = NULL;
2383 input->version = NULL;
2384 input->consumed = 0;
2385 input->length = 0;
2386 return(input);
2387}
2388#endif
2389
2390
2391/************************************************************************
2392 * *
2393 * Commodity functions, cleanup needed ? *
2394 * *
2395 ************************************************************************/
2396/*
2397 * all tags allowing pc data from the html 4.01 loose dtd
2398 * NOTE: it might be more appropriate to integrate this information
2399 * into the html40ElementTable array but I don't want to risk any
2400 * binary incompatibility
2401 */
2402static const char *allowPCData[] = {
2403 "a", "abbr", "acronym", "address", "applet", "b", "bdo", "big",
2404 "blockquote", "body", "button", "caption", "center", "cite", "code",
2405 "dd", "del", "dfn", "div", "dt", "em", "font", "form", "h1", "h2",
2406 "h3", "h4", "h5", "h6", "i", "iframe", "ins", "kbd", "label", "legend",
2407 "li", "noframes", "noscript", "object", "p", "pre", "q", "s", "samp",
2408 "small", "span", "strike", "strong", "td", "th", "tt", "u", "var"
2409};
2410
2411/**
2412 * areBlanks:
2413 * @ctxt: an HTML parser context
2414 * @str: a xmlChar *
2415 * @len: the size of @str
2416 *
2417 * Is this a sequence of blank chars that one can ignore ?
2418 *
2419 * Returns 1 if ignorable 0 otherwise.
2420 */
2421
2422static int areBlanks(htmlParserCtxtPtr ctxt, const xmlChar *str, int len) {
2423 unsigned int i;
2424 int j;
2425 xmlNodePtr lastChild;
2426 xmlDtdPtr dtd;
2427
2428 for (j = 0;j < len;j++)
2429 if (!(IS_BLANK_CH(str[j]))) return(0);
2430
2431 if (CUR == 0) return(1);
2432 if (CUR != '<') return(0);
2433 if (ctxt->name == NULL)
2434 return(1);
2435 if (xmlStrEqual(ctxt->name, BAD_CAST"html"))
2436 return(1);
2437 if (xmlStrEqual(ctxt->name, BAD_CAST"head"))
2438 return(1);
2439
2440 /* Only strip CDATA children of the body tag for strict HTML DTDs */
2441 if (xmlStrEqual(ctxt->name, BAD_CAST "body") && ctxt->myDoc != NULL) {
2442 dtd = xmlGetIntSubset(ctxt->myDoc);
2443 if (dtd != NULL && dtd->ExternalID != NULL) {
2444 if (!xmlStrcasecmp(dtd->ExternalID, BAD_CAST "-//W3C//DTD HTML 4.01//EN") ||
2445 !xmlStrcasecmp(dtd->ExternalID, BAD_CAST "-//W3C//DTD HTML 4//EN"))
2446 return(1);
2447 }
2448 }
2449
2450 if (ctxt->node == NULL) return(0);
2451 lastChild = xmlGetLastChild(ctxt->node);
2452 while ((lastChild) && (lastChild->type == XML_COMMENT_NODE))
2453 lastChild = lastChild->prev;
2454 if (lastChild == NULL) {
2455 if ((ctxt->node->type != XML_ELEMENT_NODE) &&
2456 (ctxt->node->content != NULL)) return(0);
2457 /* keep ws in constructs like ...<b> </b>...
2458 for all tags "b" allowing PCDATA */
2459 for ( i = 0; i < sizeof(allowPCData)/sizeof(allowPCData[0]); i++ ) {
2460 if ( xmlStrEqual(ctxt->name, BAD_CAST allowPCData[i]) ) {
2461 return(0);
2462 }
2463 }
2464 } else if (xmlNodeIsText(lastChild)) {
2465 return(0);
2466 } else {
2467 /* keep ws in constructs like <p><b>xy</b> <i>z</i><p>
2468 for all tags "p" allowing PCDATA */
2469 for ( i = 0; i < sizeof(allowPCData)/sizeof(allowPCData[0]); i++ ) {
2470 if ( xmlStrEqual(lastChild->name, BAD_CAST allowPCData[i]) ) {
2471 return(0);
2472 }
2473 }
2474 }
2475 return(1);
2476}
2477
2478/**
2479 * htmlNewDocNoDtD:
2480 * @URI: URI for the dtd, or NULL
2481 * @ExternalID: the external ID of the DTD, or NULL
2482 *
2483 * Creates a new HTML document without a DTD node if @URI and @ExternalID
2484 * are NULL
2485 *
2486 * Returns a new document, do not initialize the DTD if not provided
2487 */
2488htmlDocPtr
2489htmlNewDocNoDtD(const xmlChar *URI, const xmlChar *ExternalID) {
2490 xmlDocPtr cur;
2491
2492 /*
2493 * Allocate a new document and fill the fields.
2494 */
2495 cur = (xmlDocPtr) xmlMalloc(sizeof(xmlDoc));
2496 if (cur == NULL) {
2497 htmlErrMemory(NULL, "HTML document creation failed\n");
2498 return(NULL);
2499 }
2500 memset(cur, 0, sizeof(xmlDoc));
2501
2502 cur->type = XML_HTML_DOCUMENT_NODE;
2503 cur->version = NULL;
2504 cur->intSubset = NULL;
2505 cur->doc = cur;
2506 cur->name = NULL;
2507 cur->children = NULL;
2508 cur->extSubset = NULL;
2509 cur->oldNs = NULL;
2510 cur->encoding = NULL;
2511 cur->standalone = 1;
2512 cur->compression = 0;
2513 cur->ids = NULL;
2514 cur->refs = NULL;
2515 cur->_private = NULL;
2516 cur->charset = XML_CHAR_ENCODING_UTF8;
2517 cur->properties = XML_DOC_HTML | XML_DOC_USERBUILT;
2518 if ((ExternalID != NULL) ||
2519 (URI != NULL))
2520 xmlCreateIntSubset(cur, BAD_CAST "html", ExternalID, URI);
2521 if ((__xmlRegisterCallbacks) && (xmlRegisterNodeDefaultValue))
2522 xmlRegisterNodeDefaultValue((xmlNodePtr)cur);
2523 return(cur);
2524}
2525
2526/**
2527 * htmlNewDoc:
2528 * @URI: URI for the dtd, or NULL
2529 * @ExternalID: the external ID of the DTD, or NULL
2530 *
2531 * Creates a new HTML document
2532 *
2533 * Returns a new document
2534 */
2535htmlDocPtr
2536htmlNewDoc(const xmlChar *URI, const xmlChar *ExternalID) {
2537 if ((URI == NULL) && (ExternalID == NULL))
2538 return(htmlNewDocNoDtD(
2539 BAD_CAST "http://www.w3.org/TR/REC-html40/loose.dtd",
2540 BAD_CAST "-//W3C//DTD HTML 4.0 Transitional//EN"));
2541
2542 return(htmlNewDocNoDtD(URI, ExternalID));
2543}
2544
2545
2546/************************************************************************
2547 * *
2548 * The parser itself *
2549 * Relates to http://www.w3.org/TR/html40 *
2550 * *
2551 ************************************************************************/
2552
2553/************************************************************************
2554 * *
2555 * The parser itself *
2556 * *
2557 ************************************************************************/
2558
2559static const xmlChar * htmlParseNameComplex(xmlParserCtxtPtr ctxt);
2560
2561/**
2562 * htmlParseHTMLName:
2563 * @ctxt: an HTML parser context
2564 *
2565 * parse an HTML tag or attribute name, note that we convert it to lowercase
2566 * since HTML names are not case-sensitive.
2567 *
2568 * Returns the Tag Name parsed or NULL
2569 */
2570
2571static const xmlChar *
2572htmlParseHTMLName(htmlParserCtxtPtr ctxt) {
2573 int i = 0;
2574 xmlChar loc[HTML_PARSER_BUFFER_SIZE];
2575
2576 if (!IS_ASCII_LETTER(CUR) && (CUR != '_') &&
2577 (CUR != ':') && (CUR != '.')) return(NULL);
2578
2579 while ((i < HTML_PARSER_BUFFER_SIZE) &&
2580 ((IS_ASCII_LETTER(CUR)) || (IS_ASCII_DIGIT(CUR)) ||
2581 (CUR == ':') || (CUR == '-') || (CUR == '_') ||
2582 (CUR == '.'))) {
2583 if ((CUR >= 'A') && (CUR <= 'Z')) loc[i] = CUR + 0x20;
2584 else loc[i] = CUR;
2585 i++;
2586
2587 NEXT;
2588 }
2589
2590 return(xmlDictLookup(ctxt->dict, loc, i));
2591}
2592
2593
2594/**
2595 * htmlParseHTMLName_nonInvasive:
2596 * @ctxt: an HTML parser context
2597 *
2598 * parse an HTML tag or attribute name, note that we convert it to lowercase
2599 * since HTML names are not case-sensitive, this doesn't consume the data
2600 * from the stream, it's a look-ahead
2601 *
2602 * Returns the Tag Name parsed or NULL
2603 */
2604
2605static const xmlChar *
2606htmlParseHTMLName_nonInvasive(htmlParserCtxtPtr ctxt) {
2607 int i = 0;
2608 xmlChar loc[HTML_PARSER_BUFFER_SIZE];
2609
2610 if (!IS_ASCII_LETTER(NXT(1)) && (NXT(1) != '_') &&
2611 (NXT(1) != ':')) return(NULL);
2612
2613 while ((i < HTML_PARSER_BUFFER_SIZE) &&
2614 ((IS_ASCII_LETTER(NXT(1+i))) || (IS_ASCII_DIGIT(NXT(1+i))) ||
2615 (NXT(1+i) == ':') || (NXT(1+i) == '-') || (NXT(1+i) == '_'))) {
2616 if ((NXT(1+i) >= 'A') && (NXT(1+i) <= 'Z')) loc[i] = NXT(1+i) + 0x20;
2617 else loc[i] = NXT(1+i);
2618 i++;
2619 }
2620
2621 return(xmlDictLookup(ctxt->dict, loc, i));
2622}
2623
2624
2625/**
2626 * htmlParseName:
2627 * @ctxt: an HTML parser context
2628 *
2629 * parse an HTML name, this routine is case sensitive.
2630 *
2631 * Returns the Name parsed or NULL
2632 */
2633
2634static const xmlChar *
2635htmlParseName(htmlParserCtxtPtr ctxt) {
2636 const xmlChar *in;
2637 const xmlChar *ret;
2638 int count = 0;
2639
2640 GROW;
2641
2642 /*
2643 * Accelerator for simple ASCII names
2644 */
2645 in = ctxt->input->cur;
2646 if (((*in >= 0x61) && (*in <= 0x7A)) ||
2647 ((*in >= 0x41) && (*in <= 0x5A)) ||
2648 (*in == '_') || (*in == ':')) {
2649 in++;
2650 while (((*in >= 0x61) && (*in <= 0x7A)) ||
2651 ((*in >= 0x41) && (*in <= 0x5A)) ||
2652 ((*in >= 0x30) && (*in <= 0x39)) ||
2653 (*in == '_') || (*in == '-') ||
2654 (*in == ':') || (*in == '.'))
2655 in++;
2656
2657 if (in == ctxt->input->end)
2658 return(NULL);
2659
2660 if ((*in > 0) && (*in < 0x80)) {
2661 count = in - ctxt->input->cur;
2662 ret = xmlDictLookup(ctxt->dict, ctxt->input->cur, count);
2663 ctxt->input->cur = in;
2664 ctxt->input->col += count;
2665 return(ret);
2666 }
2667 }
2668 return(htmlParseNameComplex(ctxt));
2669}
2670
2671static const xmlChar *
2672htmlParseNameComplex(xmlParserCtxtPtr ctxt) {
2673 int len = 0, l;
2674 int c;
2675 int count = 0;
2676 const xmlChar *base = ctxt->input->base;
2677
2678 /*
2679 * Handler for more complex cases
2680 */
2681 GROW;
2682 c = CUR_CHAR(l);
2683 if ((c == ' ') || (c == '>') || (c == '/') || /* accelerators */
2684 (!IS_LETTER(c) && (c != '_') &&
2685 (c != ':'))) {
2686 return(NULL);
2687 }
2688
2689 while ((c != ' ') && (c != '>') && (c != '/') && /* test bigname.xml */
2690 ((IS_LETTER(c)) || (IS_DIGIT(c)) ||
2691 (c == '.') || (c == '-') ||
2692 (c == '_') || (c == ':') ||
2693 (IS_COMBINING(c)) ||
2694 (IS_EXTENDER(c)))) {
2695 if (count++ > 100) {
2696 count = 0;
2697 GROW;
2698 }
2699 len += l;
2700 NEXTL(l);
2701 c = CUR_CHAR(l);
2702 if (ctxt->input->base != base) {
2703 /*
2704 * We changed encoding from an unknown encoding
2705 * Input buffer changed location, so we better start again
2706 */
2707 return(htmlParseNameComplex(ctxt));
2708 }
2709 }
2710
2711 if (ctxt->input->cur - ctxt->input->base < len) {
2712 /* Sanity check */
2713 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
2714 "unexpected change of input buffer", NULL, NULL);
2715 return (NULL);
2716 }
2717
2718 return(xmlDictLookup(ctxt->dict, ctxt->input->cur - len, len));
2719}
2720
2721
2722/**
2723 * htmlParseHTMLAttribute:
2724 * @ctxt: an HTML parser context
2725 * @stop: a char stop value
2726 *
2727 * parse an HTML attribute value till the stop (quote), if
2728 * stop is 0 then it stops at the first space
2729 *
2730 * Returns the attribute parsed or NULL
2731 */
2732
2733static xmlChar *
2734htmlParseHTMLAttribute(htmlParserCtxtPtr ctxt, const xmlChar stop) {
2735 xmlChar *buffer = NULL;
2736 int buffer_size = 0;
2737 xmlChar *out = NULL;
2738 const xmlChar *name = NULL;
2739 const xmlChar *cur = NULL;
2740 const htmlEntityDesc * ent;
2741
2742 /*
2743 * allocate a translation buffer.
2744 */
2745 buffer_size = HTML_PARSER_BUFFER_SIZE;
2746 buffer = (xmlChar *) xmlMallocAtomic(buffer_size * sizeof(xmlChar));
2747 if (buffer == NULL) {
2748 htmlErrMemory(ctxt, "buffer allocation failed\n");
2749 return(NULL);
2750 }
2751 out = buffer;
2752
2753 /*
2754 * Ok loop until we reach one of the ending chars
2755 */
2756 while ((CUR != 0) && (CUR != stop)) {
2757 if ((stop == 0) && (CUR == '>')) break;
2758 if ((stop == 0) && (IS_BLANK_CH(CUR))) break;
2759 if (CUR == '&') {
2760 if (NXT(1) == '#') {
2761 unsigned int c;
2762 int bits;
2763
2764 c = htmlParseCharRef(ctxt);
2765 if (c < 0x80)
2766 { *out++ = c; bits= -6; }
2767 else if (c < 0x800)
2768 { *out++ =((c >> 6) & 0x1F) | 0xC0; bits= 0; }
2769 else if (c < 0x10000)
2770 { *out++ =((c >> 12) & 0x0F) | 0xE0; bits= 6; }
2771 else
2772 { *out++ =((c >> 18) & 0x07) | 0xF0; bits= 12; }
2773
2774 for ( ; bits >= 0; bits-= 6) {
2775 *out++ = ((c >> bits) & 0x3F) | 0x80;
2776 }
2777
2778 if (out - buffer > buffer_size - 100) {
2779 int indx = out - buffer;
2780
2781 growBuffer(buffer);
2782 out = &buffer[indx];
2783 }
2784 } else {
2785 ent = htmlParseEntityRef(ctxt, &name);
2786 if (name == NULL) {
2787 *out++ = '&';
2788 if (out - buffer > buffer_size - 100) {
2789 int indx = out - buffer;
2790
2791 growBuffer(buffer);
2792 out = &buffer[indx];
2793 }
2794 } else if (ent == NULL) {
2795 *out++ = '&';
2796 cur = name;
2797 while (*cur != 0) {
2798 if (out - buffer > buffer_size - 100) {
2799 int indx = out - buffer;
2800
2801 growBuffer(buffer);
2802 out = &buffer[indx];
2803 }
2804 *out++ = *cur++;
2805 }
2806 } else {
2807 unsigned int c;
2808 int bits;
2809
2810 if (out - buffer > buffer_size - 100) {
2811 int indx = out - buffer;
2812
2813 growBuffer(buffer);
2814 out = &buffer[indx];
2815 }
2816 c = ent->value;
2817 if (c < 0x80)
2818 { *out++ = c; bits= -6; }
2819 else if (c < 0x800)
2820 { *out++ =((c >> 6) & 0x1F) | 0xC0; bits= 0; }
2821 else if (c < 0x10000)
2822 { *out++ =((c >> 12) & 0x0F) | 0xE0; bits= 6; }
2823 else
2824 { *out++ =((c >> 18) & 0x07) | 0xF0; bits= 12; }
2825
2826 for ( ; bits >= 0; bits-= 6) {
2827 *out++ = ((c >> bits) & 0x3F) | 0x80;
2828 }
2829 }
2830 }
2831 } else {
2832 unsigned int c;
2833 int bits, l;
2834
2835 if (out - buffer > buffer_size - 100) {
2836 int indx = out - buffer;
2837
2838 growBuffer(buffer);
2839 out = &buffer[indx];
2840 }
2841 c = CUR_CHAR(l);
2842 if (c < 0x80)
2843 { *out++ = c; bits= -6; }
2844 else if (c < 0x800)
2845 { *out++ =((c >> 6) & 0x1F) | 0xC0; bits= 0; }
2846 else if (c < 0x10000)
2847 { *out++ =((c >> 12) & 0x0F) | 0xE0; bits= 6; }
2848 else
2849 { *out++ =((c >> 18) & 0x07) | 0xF0; bits= 12; }
2850
2851 for ( ; bits >= 0; bits-= 6) {
2852 *out++ = ((c >> bits) & 0x3F) | 0x80;
2853 }
2854 NEXT;
2855 }
2856 }
2857 *out = 0;
2858 return(buffer);
2859}
2860
2861/**
2862 * htmlParseEntityRef:
2863 * @ctxt: an HTML parser context
2864 * @str: location to store the entity name
2865 *
2866 * parse an HTML ENTITY references
2867 *
2868 * [68] EntityRef ::= '&' Name ';'
2869 *
2870 * Returns the associated htmlEntityDescPtr if found, or NULL otherwise,
2871 * if non-NULL *str will have to be freed by the caller.
2872 */
2873const htmlEntityDesc *
2874htmlParseEntityRef(htmlParserCtxtPtr ctxt, const xmlChar **str) {
2875 const xmlChar *name;
2876 const htmlEntityDesc * ent = NULL;
2877
2878 if (str != NULL) *str = NULL;
2879 if ((ctxt == NULL) || (ctxt->input == NULL)) return(NULL);
2880
2881 if (CUR == '&') {
2882 NEXT;
2883 name = htmlParseName(ctxt);
2884 if (name == NULL) {
2885 htmlParseErr(ctxt, XML_ERR_NAME_REQUIRED,
2886 "htmlParseEntityRef: no name\n", NULL, NULL);
2887 } else {
2888 GROW;
2889 if (CUR == ';') {
2890 if (str != NULL)
2891 *str = name;
2892
2893 /*
2894 * Lookup the entity in the table.
2895 */
2896 ent = htmlEntityLookup(name);
2897 if (ent != NULL) /* OK that's ugly !!! */
2898 NEXT;
2899 } else {
2900 htmlParseErr(ctxt, XML_ERR_ENTITYREF_SEMICOL_MISSING,
2901 "htmlParseEntityRef: expecting ';'\n",
2902 NULL, NULL);
2903 if (str != NULL)
2904 *str = name;
2905 }
2906 }
2907 }
2908 return(ent);
2909}
2910
2911/**
2912 * htmlParseAttValue:
2913 * @ctxt: an HTML parser context
2914 *
2915 * parse a value for an attribute
2916 * Note: the parser won't do substitution of entities here, this
2917 * will be handled later in xmlStringGetNodeList, unless it was
2918 * asked for ctxt->replaceEntities != 0
2919 *
2920 * Returns the AttValue parsed or NULL.
2921 */
2922
2923static xmlChar *
2924htmlParseAttValue(htmlParserCtxtPtr ctxt) {
2925 xmlChar *ret = NULL;
2926
2927 if (CUR == '"') {
2928 NEXT;
2929 ret = htmlParseHTMLAttribute(ctxt, '"');
2930 if (CUR != '"') {
2931 htmlParseErr(ctxt, XML_ERR_ATTRIBUTE_NOT_FINISHED,
2932 "AttValue: \" expected\n", NULL, NULL);
2933 } else
2934 NEXT;
2935 } else if (CUR == '\'') {
2936 NEXT;
2937 ret = htmlParseHTMLAttribute(ctxt, '\'');
2938 if (CUR != '\'') {
2939 htmlParseErr(ctxt, XML_ERR_ATTRIBUTE_NOT_FINISHED,
2940 "AttValue: ' expected\n", NULL, NULL);
2941 } else
2942 NEXT;
2943 } else {
2944 /*
2945 * That's an HTMLism, the attribute value may not be quoted
2946 */
2947 ret = htmlParseHTMLAttribute(ctxt, 0);
2948 if (ret == NULL) {
2949 htmlParseErr(ctxt, XML_ERR_ATTRIBUTE_WITHOUT_VALUE,
2950 "AttValue: no value found\n", NULL, NULL);
2951 }
2952 }
2953 return(ret);
2954}
2955
2956/**
2957 * htmlParseSystemLiteral:
2958 * @ctxt: an HTML parser context
2959 *
2960 * parse an HTML Literal
2961 *
2962 * [11] SystemLiteral ::= ('"' [^"]* '"') | ("'" [^']* "'")
2963 *
2964 * Returns the SystemLiteral parsed or NULL
2965 */
2966
2967static xmlChar *
2968htmlParseSystemLiteral(htmlParserCtxtPtr ctxt) {
2969 size_t len = 0, startPosition = 0;
2970 int err = 0;
2971 int quote;
2972 xmlChar *ret = NULL;
2973
2974 if ((CUR != '"') && (CUR != '\'')) {
2975 htmlParseErr(ctxt, XML_ERR_LITERAL_NOT_STARTED,
2976 "SystemLiteral \" or ' expected\n", NULL, NULL);
2977 return(NULL);
2978 }
2979 quote = CUR;
2980 NEXT;
2981
2982 if (CUR_PTR < BASE_PTR)
2983 return(ret);
2984 startPosition = CUR_PTR - BASE_PTR;
2985
2986 while ((CUR != 0) && (CUR != quote)) {
2987 /* TODO: Handle UTF-8 */
2988 if (!IS_CHAR_CH(CUR)) {
2989 htmlParseErrInt(ctxt, XML_ERR_INVALID_CHAR,
2990 "Invalid char in SystemLiteral 0x%X\n", CUR);
2991 err = 1;
2992 }
2993 NEXT;
2994 len++;
2995 }
2996 if (CUR != quote) {
2997 htmlParseErr(ctxt, XML_ERR_LITERAL_NOT_FINISHED,
2998 "Unfinished SystemLiteral\n", NULL, NULL);
2999 } else {
3000 NEXT;
3001 if (err == 0)
3002 ret = xmlStrndup((BASE_PTR+startPosition), len);
3003 }
3004
3005 return(ret);
3006}
3007
3008/**
3009 * htmlParsePubidLiteral:
3010 * @ctxt: an HTML parser context
3011 *
3012 * parse an HTML public literal
3013 *
3014 * [12] PubidLiteral ::= '"' PubidChar* '"' | "'" (PubidChar - "'")* "'"
3015 *
3016 * Returns the PubidLiteral parsed or NULL.
3017 */
3018
3019static xmlChar *
3020htmlParsePubidLiteral(htmlParserCtxtPtr ctxt) {
3021 size_t len = 0, startPosition = 0;
3022 int err = 0;
3023 int quote;
3024 xmlChar *ret = NULL;
3025
3026 if ((CUR != '"') && (CUR != '\'')) {
3027 htmlParseErr(ctxt, XML_ERR_LITERAL_NOT_STARTED,
3028 "PubidLiteral \" or ' expected\n", NULL, NULL);
3029 return(NULL);
3030 }
3031 quote = CUR;
3032 NEXT;
3033
3034 /*
3035 * Name ::= (Letter | '_') (NameChar)*
3036 */
3037 if (CUR_PTR < BASE_PTR)
3038 return(ret);
3039 startPosition = CUR_PTR - BASE_PTR;
3040
3041 while ((CUR != 0) && (CUR != quote)) {
3042 if (!IS_PUBIDCHAR_CH(CUR)) {
3043 htmlParseErrInt(ctxt, XML_ERR_INVALID_CHAR,
3044 "Invalid char in PubidLiteral 0x%X\n", CUR);
3045 err = 1;
3046 }
3047 len++;
3048 NEXT;
3049 }
3050
3051 if (CUR != quote) {
3052 htmlParseErr(ctxt, XML_ERR_LITERAL_NOT_FINISHED,
3053 "Unfinished PubidLiteral\n", NULL, NULL);
3054 } else {
3055 NEXT;
3056 if (err == 0)
3057 ret = xmlStrndup((BASE_PTR + startPosition), len);
3058 }
3059
3060 return(ret);
3061}
3062
3063/**
3064 * htmlParseScript:
3065 * @ctxt: an HTML parser context
3066 *
3067 * parse the content of an HTML SCRIPT or STYLE element
3068 * http://www.w3.org/TR/html4/sgml/dtd.html#Script
3069 * http://www.w3.org/TR/html4/sgml/dtd.html#StyleSheet
3070 * http://www.w3.org/TR/html4/types.html#type-script
3071 * http://www.w3.org/TR/html4/types.html#h-6.15
3072 * http://www.w3.org/TR/html4/appendix/notes.html#h-B.3.2.1
3073 *
3074 * Script data ( %Script; in the DTD) can be the content of the SCRIPT
3075 * element and the value of intrinsic event attributes. User agents must
3076 * not evaluate script data as HTML markup but instead must pass it on as
3077 * data to a script engine.
3078 * NOTES:
3079 * - The content is passed like CDATA
3080 * - the attributes for style and scripting "onXXX" are also described
3081 * as CDATA but SGML allows entities references in attributes so their
3082 * processing is identical as other attributes
3083 */
3084static void
3085htmlParseScript(htmlParserCtxtPtr ctxt) {
3086 xmlChar buf[HTML_PARSER_BIG_BUFFER_SIZE + 5];
3087 int nbchar = 0;
3088 int cur,l;
3089
3090 SHRINK;
3091 cur = CUR_CHAR(l);
3092 while (cur != 0) {
3093 if ((cur == '<') && (NXT(1) == '/')) {
3094 /*
3095 * One should break here, the specification is clear:
3096 * Authors should therefore escape "</" within the content.
3097 * Escape mechanisms are specific to each scripting or
3098 * style sheet language.
3099 *
3100 * In recovery mode, only break if end tag match the
3101 * current tag, effectively ignoring all tags inside the
3102 * script/style block and treating the entire block as
3103 * CDATA.
3104 */
3105 if (ctxt->recovery) {
3106 if (xmlStrncasecmp(ctxt->name, ctxt->input->cur+2,
3107 xmlStrlen(ctxt->name)) == 0)
3108 {
3109 break; /* while */
3110 } else {
3111 htmlParseErr(ctxt, XML_ERR_TAG_NAME_MISMATCH,
3112 "Element %s embeds close tag\n",
3113 ctxt->name, NULL);
3114 }
3115 } else {
3116 if (((NXT(2) >= 'A') && (NXT(2) <= 'Z')) ||
3117 ((NXT(2) >= 'a') && (NXT(2) <= 'z')))
3118 {
3119 break; /* while */
3120 }
3121 }
3122 }
3123 if (IS_CHAR(cur)) {
3124 COPY_BUF(l,buf,nbchar,cur);
3125 } else {
3126 htmlParseErrInt(ctxt, XML_ERR_INVALID_CHAR,
3127 "Invalid char in CDATA 0x%X\n", cur);
3128 }
3129 if (nbchar >= HTML_PARSER_BIG_BUFFER_SIZE) {
3130 buf[nbchar] = 0;
3131 if (ctxt->sax->cdataBlock!= NULL) {
3132 /*
3133 * Insert as CDATA, which is the same as HTML_PRESERVE_NODE
3134 */
3135 ctxt->sax->cdataBlock(ctxt->userData, buf, nbchar);
3136 } else if (ctxt->sax->characters != NULL) {
3137 ctxt->sax->characters(ctxt->userData, buf, nbchar);
3138 }
3139 nbchar = 0;
3140 }
3141 GROW;
3142 NEXTL(l);
3143 cur = CUR_CHAR(l);
3144 }
3145
3146 if ((nbchar != 0) && (ctxt->sax != NULL) && (!ctxt->disableSAX)) {
3147 buf[nbchar] = 0;
3148 if (ctxt->sax->cdataBlock!= NULL) {
3149 /*
3150 * Insert as CDATA, which is the same as HTML_PRESERVE_NODE
3151 */
3152 ctxt->sax->cdataBlock(ctxt->userData, buf, nbchar);
3153 } else if (ctxt->sax->characters != NULL) {
3154 ctxt->sax->characters(ctxt->userData, buf, nbchar);
3155 }
3156 }
3157}
3158
3159
3160/**
3161 * htmlParseCharDataInternal:
3162 * @ctxt: an HTML parser context
3163 * @readahead: optional read ahead character in ascii range
3164 *
3165 * parse a CharData section.
3166 * if we are within a CDATA section ']]>' marks an end of section.
3167 *
3168 * [14] CharData ::= [^<&]* - ([^<&]* ']]>' [^<&]*)
3169 */
3170
3171static void
3172htmlParseCharDataInternal(htmlParserCtxtPtr ctxt, int readahead) {
3173 xmlChar buf[HTML_PARSER_BIG_BUFFER_SIZE + 6];
3174 int nbchar = 0;
3175 int cur, l;
3176 int chunk = 0;
3177
3178 if (readahead)
3179 buf[nbchar++] = readahead;
3180
3181 SHRINK;
3182 cur = CUR_CHAR(l);
3183 while (((cur != '<') || (ctxt->token == '<')) &&
3184 ((cur != '&') || (ctxt->token == '&')) &&
3185 (cur != 0)) {
3186 if (!(IS_CHAR(cur))) {
3187 htmlParseErrInt(ctxt, XML_ERR_INVALID_CHAR,
3188 "Invalid char in CDATA 0x%X\n", cur);
3189 } else {
3190 COPY_BUF(l,buf,nbchar,cur);
3191 }
3192 if (nbchar >= HTML_PARSER_BIG_BUFFER_SIZE) {
3193 buf[nbchar] = 0;
3194
3195 /*
3196 * Ok the segment is to be consumed as chars.
3197 */
3198 if ((ctxt->sax != NULL) && (!ctxt->disableSAX)) {
3199 if (areBlanks(ctxt, buf, nbchar)) {
3200 if (ctxt->keepBlanks) {
3201 if (ctxt->sax->characters != NULL)
3202 ctxt->sax->characters(ctxt->userData, buf, nbchar);
3203 } else {
3204 if (ctxt->sax->ignorableWhitespace != NULL)
3205 ctxt->sax->ignorableWhitespace(ctxt->userData,
3206 buf, nbchar);
3207 }
3208 } else {
3209 htmlCheckParagraph(ctxt);
3210 if (ctxt->sax->characters != NULL)
3211 ctxt->sax->characters(ctxt->userData, buf, nbchar);
3212 }
3213 }
3214 nbchar = 0;
3215 }
3216 NEXTL(l);
3217 chunk++;
3218 if (chunk > HTML_PARSER_BUFFER_SIZE) {
3219 chunk = 0;
3220 SHRINK;
3221 GROW;
3222 }
3223 cur = CUR_CHAR(l);
3224 if (cur == 0) {
3225 SHRINK;
3226 GROW;
3227 cur = CUR_CHAR(l);
3228 }
3229 }
3230 if (nbchar != 0) {
3231 buf[nbchar] = 0;
3232
3233 /*
3234 * Ok the segment is to be consumed as chars.
3235 */
3236 if ((ctxt->sax != NULL) && (!ctxt->disableSAX)) {
3237 if (areBlanks(ctxt, buf, nbchar)) {
3238 if (ctxt->keepBlanks) {
3239 if (ctxt->sax->characters != NULL)
3240 ctxt->sax->characters(ctxt->userData, buf, nbchar);
3241 } else {
3242 if (ctxt->sax->ignorableWhitespace != NULL)
3243 ctxt->sax->ignorableWhitespace(ctxt->userData,
3244 buf, nbchar);
3245 }
3246 } else {
3247 htmlCheckParagraph(ctxt);
3248 if (ctxt->sax->characters != NULL)
3249 ctxt->sax->characters(ctxt->userData, buf, nbchar);
3250 }
3251 }
3252 } else {
3253 /*
3254 * Loop detection
3255 */
3256 if (cur == 0)
3257 ctxt->instate = XML_PARSER_EOF;
3258 }
3259}
3260
3261/**
3262 * htmlParseCharData:
3263 * @ctxt: an HTML parser context
3264 *
3265 * parse a CharData section.
3266 * if we are within a CDATA section ']]>' marks an end of section.
3267 *
3268 * [14] CharData ::= [^<&]* - ([^<&]* ']]>' [^<&]*)
3269 */
3270
3271static void
3272htmlParseCharData(htmlParserCtxtPtr ctxt) {
3273 htmlParseCharDataInternal(ctxt, 0);
3274}
3275
3276/**
3277 * htmlParseExternalID:
3278 * @ctxt: an HTML parser context
3279 * @publicID: a xmlChar** receiving PubidLiteral
3280 *
3281 * Parse an External ID or a Public ID
3282 *
3283 * [75] ExternalID ::= 'SYSTEM' S SystemLiteral
3284 * | 'PUBLIC' S PubidLiteral S SystemLiteral
3285 *
3286 * [83] PublicID ::= 'PUBLIC' S PubidLiteral
3287 *
3288 * Returns the function returns SystemLiteral and in the second
3289 * case publicID receives PubidLiteral, is strict is off
3290 * it is possible to return NULL and have publicID set.
3291 */
3292
3293static xmlChar *
3294htmlParseExternalID(htmlParserCtxtPtr ctxt, xmlChar **publicID) {
3295 xmlChar *URI = NULL;
3296
3297 if ((UPPER == 'S') && (UPP(1) == 'Y') &&
3298 (UPP(2) == 'S') && (UPP(3) == 'T') &&
3299 (UPP(4) == 'E') && (UPP(5) == 'M')) {
3300 SKIP(6);
3301 if (!IS_BLANK_CH(CUR)) {
3302 htmlParseErr(ctxt, XML_ERR_SPACE_REQUIRED,
3303 "Space required after 'SYSTEM'\n", NULL, NULL);
3304 }
3305 SKIP_BLANKS;
3306 URI = htmlParseSystemLiteral(ctxt);
3307 if (URI == NULL) {
3308 htmlParseErr(ctxt, XML_ERR_URI_REQUIRED,
3309 "htmlParseExternalID: SYSTEM, no URI\n", NULL, NULL);
3310 }
3311 } else if ((UPPER == 'P') && (UPP(1) == 'U') &&
3312 (UPP(2) == 'B') && (UPP(3) == 'L') &&
3313 (UPP(4) == 'I') && (UPP(5) == 'C')) {
3314 SKIP(6);
3315 if (!IS_BLANK_CH(CUR)) {
3316 htmlParseErr(ctxt, XML_ERR_SPACE_REQUIRED,
3317 "Space required after 'PUBLIC'\n", NULL, NULL);
3318 }
3319 SKIP_BLANKS;
3320 *publicID = htmlParsePubidLiteral(ctxt);
3321 if (*publicID == NULL) {
3322 htmlParseErr(ctxt, XML_ERR_PUBID_REQUIRED,
3323 "htmlParseExternalID: PUBLIC, no Public Identifier\n",
3324 NULL, NULL);
3325 }
3326 SKIP_BLANKS;
3327 if ((CUR == '"') || (CUR == '\'')) {
3328 URI = htmlParseSystemLiteral(ctxt);
3329 }
3330 }
3331 return(URI);
3332}
3333
3334/**
3335 * xmlParsePI:
3336 * @ctxt: an XML parser context
3337 *
3338 * parse an XML Processing Instruction.
3339 *
3340 * [16] PI ::= '<?' PITarget (S (Char* - (Char* '?>' Char*)))? '?>'
3341 */
3342static void
3343htmlParsePI(htmlParserCtxtPtr ctxt) {
3344 xmlChar *buf = NULL;
3345 int len = 0;
3346 int size = HTML_PARSER_BUFFER_SIZE;
3347 int cur, l;
3348 const xmlChar *target;
3349 xmlParserInputState state;
3350 int count = 0;
3351
3352 if ((RAW == '<') && (NXT(1) == '?')) {
3353 state = ctxt->instate;
3354 ctxt->instate = XML_PARSER_PI;
3355 /*
3356 * this is a Processing Instruction.
3357 */
3358 SKIP(2);
3359 SHRINK;
3360
3361 /*
3362 * Parse the target name and check for special support like
3363 * namespace.
3364 */
3365 target = htmlParseName(ctxt);
3366 if (target != NULL) {
3367 if (RAW == '>') {
3368 SKIP(1);
3369
3370 /*
3371 * SAX: PI detected.
3372 */
3373 if ((ctxt->sax) && (!ctxt->disableSAX) &&
3374 (ctxt->sax->processingInstruction != NULL))
3375 ctxt->sax->processingInstruction(ctxt->userData,
3376 target, NULL);
3377 ctxt->instate = state;
3378 return;
3379 }
3380 buf = (xmlChar *) xmlMallocAtomic(size * sizeof(xmlChar));
3381 if (buf == NULL) {
3382 htmlErrMemory(ctxt, NULL);
3383 ctxt->instate = state;
3384 return;
3385 }
3386 cur = CUR;
3387 if (!IS_BLANK(cur)) {
3388 htmlParseErr(ctxt, XML_ERR_SPACE_REQUIRED,
3389 "ParsePI: PI %s space expected\n", target, NULL);
3390 }
3391 SKIP_BLANKS;
3392 cur = CUR_CHAR(l);
3393 while ((cur != 0) && (cur != '>')) {
3394 if (len + 5 >= size) {
3395 xmlChar *tmp;
3396
3397 size *= 2;
3398 tmp = (xmlChar *) xmlRealloc(buf, size * sizeof(xmlChar));
3399 if (tmp == NULL) {
3400 htmlErrMemory(ctxt, NULL);
3401 xmlFree(buf);
3402 ctxt->instate = state;
3403 return;
3404 }
3405 buf = tmp;
3406 }
3407 count++;
3408 if (count > 50) {
3409 GROW;
3410 count = 0;
3411 }
3412 if (IS_CHAR(cur)) {
3413 COPY_BUF(l,buf,len,cur);
3414 } else {
3415 htmlParseErrInt(ctxt, XML_ERR_INVALID_CHAR,
3416 "Invalid char in processing instruction "
3417 "0x%X\n", cur);
3418 }
3419 NEXTL(l);
3420 cur = CUR_CHAR(l);
3421 if (cur == 0) {
3422 SHRINK;
3423 GROW;
3424 cur = CUR_CHAR(l);
3425 }
3426 }
3427 buf[len] = 0;
3428 if (cur != '>') {
3429 htmlParseErr(ctxt, XML_ERR_PI_NOT_FINISHED,
3430 "ParsePI: PI %s never end ...\n", target, NULL);
3431 } else {
3432 SKIP(1);
3433
3434 /*
3435 * SAX: PI detected.
3436 */
3437 if ((ctxt->sax) && (!ctxt->disableSAX) &&
3438 (ctxt->sax->processingInstruction != NULL))
3439 ctxt->sax->processingInstruction(ctxt->userData,
3440 target, buf);
3441 }
3442 xmlFree(buf);
3443 } else {
3444 htmlParseErr(ctxt, XML_ERR_PI_NOT_STARTED,
3445 "PI is not started correctly", NULL, NULL);
3446 }
3447 ctxt->instate = state;
3448 }
3449}
3450
3451/**
3452 * htmlParseComment:
3453 * @ctxt: an HTML parser context
3454 *
3455 * Parse an XML (SGML) comment <!-- .... -->
3456 *
3457 * [15] Comment ::= '<!--' ((Char - '-') | ('-' (Char - '-')))* '-->'
3458 */
3459static void
3460htmlParseComment(htmlParserCtxtPtr ctxt) {
3461 xmlChar *buf = NULL;
3462 int len;
3463 int size = HTML_PARSER_BUFFER_SIZE;
3464 int q, ql;
3465 int r, rl;
3466 int cur, l;
3467 int next, nl;
3468 xmlParserInputState state;
3469
3470 /*
3471 * Check that there is a comment right here.
3472 */
3473 if ((RAW != '<') || (NXT(1) != '!') ||
3474 (NXT(2) != '-') || (NXT(3) != '-')) return;
3475
3476 state = ctxt->instate;
3477 ctxt->instate = XML_PARSER_COMMENT;
3478 SHRINK;
3479 SKIP(4);
3480 buf = (xmlChar *) xmlMallocAtomic(size * sizeof(xmlChar));
3481 if (buf == NULL) {
3482 htmlErrMemory(ctxt, "buffer allocation failed\n");
3483 ctxt->instate = state;
3484 return;
3485 }
3486 len = 0;
3487 buf[len] = 0;
3488 q = CUR_CHAR(ql);
3489 if (q == 0)
3490 goto unfinished;
3491 NEXTL(ql);
3492 r = CUR_CHAR(rl);
3493 if (r == 0)
3494 goto unfinished;
3495 NEXTL(rl);
3496 cur = CUR_CHAR(l);
3497 while ((cur != 0) &&
3498 ((cur != '>') ||
3499 (r != '-') || (q != '-'))) {
3500 NEXTL(l);
3501 next = CUR_CHAR(nl);
3502 if (next == 0) {
3503 SHRINK;
3504 GROW;
3505 next = CUR_CHAR(nl);
3506 }
3507
3508 if ((q == '-') && (r == '-') && (cur == '!') && (next == '>')) {
3509 htmlParseErr(ctxt, XML_ERR_COMMENT_NOT_FINISHED,
3510 "Comment incorrectly closed by '--!>'", NULL, NULL);
3511 cur = '>';
3512 break;
3513 }
3514
3515 if (len + 5 >= size) {
3516 xmlChar *tmp;
3517
3518 size *= 2;
3519 tmp = (xmlChar *) xmlRealloc(buf, size * sizeof(xmlChar));
3520 if (tmp == NULL) {
3521 xmlFree(buf);
3522 htmlErrMemory(ctxt, "growing buffer failed\n");
3523 ctxt->instate = state;
3524 return;
3525 }
3526 buf = tmp;
3527 }
3528 if (IS_CHAR(q)) {
3529 COPY_BUF(ql,buf,len,q);
3530 } else {
3531 htmlParseErrInt(ctxt, XML_ERR_INVALID_CHAR,
3532 "Invalid char in comment 0x%X\n", q);
3533 }
3534
3535 q = r;
3536 ql = rl;
3537 r = cur;
3538 rl = l;
3539 cur = next;
3540 l = nl;
3541 }
3542 buf[len] = 0;
3543 if (cur == '>') {
3544 NEXT;
3545 if ((ctxt->sax != NULL) && (ctxt->sax->comment != NULL) &&
3546 (!ctxt->disableSAX))
3547 ctxt->sax->comment(ctxt->userData, buf);
3548 xmlFree(buf);
3549 ctxt->instate = state;
3550 return;
3551 }
3552
3553unfinished:
3554 htmlParseErr(ctxt, XML_ERR_COMMENT_NOT_FINISHED,
3555 "Comment not terminated \n<!--%.50s\n", buf, NULL);
3556 xmlFree(buf);
3557}
3558
3559/**
3560 * htmlParseCharRef:
3561 * @ctxt: an HTML parser context
3562 *
3563 * parse Reference declarations
3564 *
3565 * [66] CharRef ::= '&#' [0-9]+ ';' |
3566 * '&#x' [0-9a-fA-F]+ ';'
3567 *
3568 * Returns the value parsed (as an int)
3569 */
3570int
3571htmlParseCharRef(htmlParserCtxtPtr ctxt) {
3572 int val = 0;
3573
3574 if ((ctxt == NULL) || (ctxt->input == NULL)) {
3575 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
3576 "htmlParseCharRef: context error\n",
3577 NULL, NULL);
3578 return(0);
3579 }
3580 if ((CUR == '&') && (NXT(1) == '#') &&
3581 ((NXT(2) == 'x') || NXT(2) == 'X')) {
3582 SKIP(3);
3583 while (CUR != ';') {
3584 if ((CUR >= '0') && (CUR <= '9')) {
3585 if (val < 0x110000)
3586 val = val * 16 + (CUR - '0');
3587 } else if ((CUR >= 'a') && (CUR <= 'f')) {
3588 if (val < 0x110000)
3589 val = val * 16 + (CUR - 'a') + 10;
3590 } else if ((CUR >= 'A') && (CUR <= 'F')) {
3591 if (val < 0x110000)
3592 val = val * 16 + (CUR - 'A') + 10;
3593 } else {
3594 htmlParseErr(ctxt, XML_ERR_INVALID_HEX_CHARREF,
3595 "htmlParseCharRef: missing semicolon\n",
3596 NULL, NULL);
3597 break;
3598 }
3599 NEXT;
3600 }
3601 if (CUR == ';')
3602 NEXT;
3603 } else if ((CUR == '&') && (NXT(1) == '#')) {
3604 SKIP(2);
3605 while (CUR != ';') {
3606 if ((CUR >= '0') && (CUR <= '9')) {
3607 if (val < 0x110000)
3608 val = val * 10 + (CUR - '0');
3609 } else {
3610 htmlParseErr(ctxt, XML_ERR_INVALID_DEC_CHARREF,
3611 "htmlParseCharRef: missing semicolon\n",
3612 NULL, NULL);
3613 break;
3614 }
3615 NEXT;
3616 }
3617 if (CUR == ';')
3618 NEXT;
3619 } else {
3620 htmlParseErr(ctxt, XML_ERR_INVALID_CHARREF,
3621 "htmlParseCharRef: invalid value\n", NULL, NULL);
3622 }
3623 /*
3624 * Check the value IS_CHAR ...
3625 */
3626 if (IS_CHAR(val)) {
3627 return(val);
3628 } else if (val >= 0x110000) {
3629 htmlParseErr(ctxt, XML_ERR_INVALID_CHAR,
3630 "htmlParseCharRef: value too large\n", NULL, NULL);
3631 } else {
3632 htmlParseErrInt(ctxt, XML_ERR_INVALID_CHAR,
3633 "htmlParseCharRef: invalid xmlChar value %d\n",
3634 val);
3635 }
3636 return(0);
3637}
3638
3639
3640/**
3641 * htmlParseDocTypeDecl:
3642 * @ctxt: an HTML parser context
3643 *
3644 * parse a DOCTYPE declaration
3645 *
3646 * [28] doctypedecl ::= '<!DOCTYPE' S Name (S ExternalID)? S?
3647 * ('[' (markupdecl | PEReference | S)* ']' S?)? '>'
3648 */
3649
3650static void
3651htmlParseDocTypeDecl(htmlParserCtxtPtr ctxt) {
3652 const xmlChar *name;
3653 xmlChar *ExternalID = NULL;
3654 xmlChar *URI = NULL;
3655
3656 /*
3657 * We know that '<!DOCTYPE' has been detected.
3658 */
3659 SKIP(9);
3660
3661 SKIP_BLANKS;
3662
3663 /*
3664 * Parse the DOCTYPE name.
3665 */
3666 name = htmlParseName(ctxt);
3667 if (name == NULL) {
3668 htmlParseErr(ctxt, XML_ERR_NAME_REQUIRED,
3669 "htmlParseDocTypeDecl : no DOCTYPE name !\n",
3670 NULL, NULL);
3671 }
3672 /*
3673 * Check that upper(name) == "HTML" !!!!!!!!!!!!!
3674 */
3675
3676 SKIP_BLANKS;
3677
3678 /*
3679 * Check for SystemID and ExternalID
3680 */
3681 URI = htmlParseExternalID(ctxt, &ExternalID);
3682 SKIP_BLANKS;
3683
3684 /*
3685 * We should be at the end of the DOCTYPE declaration.
3686 */
3687 if (CUR != '>') {
3688 htmlParseErr(ctxt, XML_ERR_DOCTYPE_NOT_FINISHED,
3689 "DOCTYPE improperly terminated\n", NULL, NULL);
3690 /* Ignore bogus content */
3691 while ((CUR != 0) && (CUR != '>'))
3692 NEXT;
3693 }
3694 if (CUR == '>')
3695 NEXT;
3696
3697 /*
3698 * Create or update the document accordingly to the DOCTYPE
3699 */
3700 if ((ctxt->sax != NULL) && (ctxt->sax->internalSubset != NULL) &&
3701 (!ctxt->disableSAX))
3702 ctxt->sax->internalSubset(ctxt->userData, name, ExternalID, URI);
3703
3704 /*
3705 * Cleanup, since we don't use all those identifiers
3706 */
3707 if (URI != NULL) xmlFree(URI);
3708 if (ExternalID != NULL) xmlFree(ExternalID);
3709}
3710
3711/**
3712 * htmlParseAttribute:
3713 * @ctxt: an HTML parser context
3714 * @value: a xmlChar ** used to store the value of the attribute
3715 *
3716 * parse an attribute
3717 *
3718 * [41] Attribute ::= Name Eq AttValue
3719 *
3720 * [25] Eq ::= S? '=' S?
3721 *
3722 * With namespace:
3723 *
3724 * [NS 11] Attribute ::= QName Eq AttValue
3725 *
3726 * Also the case QName == xmlns:??? is handled independently as a namespace
3727 * definition.
3728 *
3729 * Returns the attribute name, and the value in *value.
3730 */
3731
3732static const xmlChar *
3733htmlParseAttribute(htmlParserCtxtPtr ctxt, xmlChar **value) {
3734 const xmlChar *name;
3735 xmlChar *val = NULL;
3736
3737 *value = NULL;
3738 name = htmlParseHTMLName(ctxt);
3739 if (name == NULL) {
3740 htmlParseErr(ctxt, XML_ERR_NAME_REQUIRED,
3741 "error parsing attribute name\n", NULL, NULL);
3742 return(NULL);
3743 }
3744
3745 /*
3746 * read the value
3747 */
3748 SKIP_BLANKS;
3749 if (CUR == '=') {
3750 NEXT;
3751 SKIP_BLANKS;
3752 val = htmlParseAttValue(ctxt);
3753 }
3754
3755 *value = val;
3756 return(name);
3757}
3758
3759/**
3760 * htmlCheckEncodingDirect:
3761 * @ctxt: an HTML parser context
3762 * @attvalue: the attribute value
3763 *
3764 * Checks an attribute value to detect
3765 * the encoding
3766 * If a new encoding is detected the parser is switched to decode
3767 * it and pass UTF8
3768 */
3769static void
3770htmlCheckEncodingDirect(htmlParserCtxtPtr ctxt, const xmlChar *encoding) {
3771
3772 if ((ctxt == NULL) || (encoding == NULL) ||
3773 (ctxt->options & HTML_PARSE_IGNORE_ENC))
3774 return;
3775
3776 /* do not change encoding */
3777 if (ctxt->input->encoding != NULL)
3778 return;
3779
3780 if (encoding != NULL) {
3781 xmlCharEncoding enc;
3782 xmlCharEncodingHandlerPtr handler;
3783
3784 while ((*encoding == ' ') || (*encoding == '\t')) encoding++;
3785
3786 if (ctxt->input->encoding != NULL)
3787 xmlFree((xmlChar *) ctxt->input->encoding);
3788 ctxt->input->encoding = xmlStrdup(encoding);
3789
3790 enc = xmlParseCharEncoding((const char *) encoding);
3791 /*
3792 * registered set of known encodings
3793 */
3794 if (enc != XML_CHAR_ENCODING_ERROR) {
3795 if (((enc == XML_CHAR_ENCODING_UTF16LE) ||
3796 (enc == XML_CHAR_ENCODING_UTF16BE) ||
3797 (enc == XML_CHAR_ENCODING_UCS4LE) ||
3798 (enc == XML_CHAR_ENCODING_UCS4BE)) &&
3799 (ctxt->input->buf != NULL) &&
3800 (ctxt->input->buf->encoder == NULL)) {
3801 htmlParseErr(ctxt, XML_ERR_INVALID_ENCODING,
3802 "htmlCheckEncoding: wrong encoding meta\n",
3803 NULL, NULL);
3804 } else {
3805 xmlSwitchEncoding(ctxt, enc);
3806 }
3807 ctxt->charset = XML_CHAR_ENCODING_UTF8;
3808 } else {
3809 /*
3810 * fallback for unknown encodings
3811 */
3812 handler = xmlFindCharEncodingHandler((const char *) encoding);
3813 if (handler != NULL) {
3814 xmlSwitchToEncoding(ctxt, handler);
3815 ctxt->charset = XML_CHAR_ENCODING_UTF8;
3816 } else {
3817 htmlParseErr(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
3818 "htmlCheckEncoding: unknown encoding %s\n",
3819 encoding, NULL);
3820 }
3821 }
3822
3823 if ((ctxt->input->buf != NULL) &&
3824 (ctxt->input->buf->encoder != NULL) &&
3825 (ctxt->input->buf->raw != NULL) &&
3826 (ctxt->input->buf->buffer != NULL)) {
3827 int nbchars;
3828 int processed;
3829
3830 /*
3831 * convert as much as possible to the parser reading buffer.
3832 */
3833 processed = ctxt->input->cur - ctxt->input->base;
3834 xmlBufShrink(ctxt->input->buf->buffer, processed);
3835 nbchars = xmlCharEncInput(ctxt->input->buf, 1);
3836 xmlBufResetInput(ctxt->input->buf->buffer, ctxt->input);
3837 if (nbchars < 0) {
3838 htmlParseErr(ctxt, XML_ERR_INVALID_ENCODING,
3839 "htmlCheckEncoding: encoder error\n",
3840 NULL, NULL);
3841 }
3842 }
3843 }
3844}
3845
3846/**
3847 * htmlCheckEncoding:
3848 * @ctxt: an HTML parser context
3849 * @attvalue: the attribute value
3850 *
3851 * Checks an http-equiv attribute from a Meta tag to detect
3852 * the encoding
3853 * If a new encoding is detected the parser is switched to decode
3854 * it and pass UTF8
3855 */
3856static void
3857htmlCheckEncoding(htmlParserCtxtPtr ctxt, const xmlChar *attvalue) {
3858 const xmlChar *encoding;
3859
3860 if (!attvalue)
3861 return;
3862
3863 encoding = xmlStrcasestr(attvalue, BAD_CAST"charset");
3864 if (encoding != NULL) {
3865 encoding += 7;
3866 }
3867 /*
3868 * skip blank
3869 */
3870 if (encoding && IS_BLANK_CH(*encoding))
3871 encoding = xmlStrcasestr(attvalue, BAD_CAST"=");
3872 if (encoding && *encoding == '=') {
3873 encoding ++;
3874 htmlCheckEncodingDirect(ctxt, encoding);
3875 }
3876}
3877
3878/**
3879 * htmlCheckMeta:
3880 * @ctxt: an HTML parser context
3881 * @atts: the attributes values
3882 *
3883 * Checks an attributes from a Meta tag
3884 */
3885static void
3886htmlCheckMeta(htmlParserCtxtPtr ctxt, const xmlChar **atts) {
3887 int i;
3888 const xmlChar *att, *value;
3889 int http = 0;
3890 const xmlChar *content = NULL;
3891
3892 if ((ctxt == NULL) || (atts == NULL))
3893 return;
3894
3895 i = 0;
3896 att = atts[i++];
3897 while (att != NULL) {
3898 value = atts[i++];
3899 if ((value != NULL) && (!xmlStrcasecmp(att, BAD_CAST"http-equiv"))
3900 && (!xmlStrcasecmp(value, BAD_CAST"Content-Type")))
3901 http = 1;
3902 else if ((value != NULL) && (!xmlStrcasecmp(att, BAD_CAST"charset")))
3903 htmlCheckEncodingDirect(ctxt, value);
3904 else if ((value != NULL) && (!xmlStrcasecmp(att, BAD_CAST"content")))
3905 content = value;
3906 att = atts[i++];
3907 }
3908 if ((http) && (content != NULL))
3909 htmlCheckEncoding(ctxt, content);
3910
3911}
3912
3913/**
3914 * htmlParseStartTag:
3915 * @ctxt: an HTML parser context
3916 *
3917 * parse a start of tag either for rule element or
3918 * EmptyElement. In both case we don't parse the tag closing chars.
3919 *
3920 * [40] STag ::= '<' Name (S Attribute)* S? '>'
3921 *
3922 * [44] EmptyElemTag ::= '<' Name (S Attribute)* S? '/>'
3923 *
3924 * With namespace:
3925 *
3926 * [NS 8] STag ::= '<' QName (S Attribute)* S? '>'
3927 *
3928 * [NS 10] EmptyElement ::= '<' QName (S Attribute)* S? '/>'
3929 *
3930 * Returns 0 in case of success, -1 in case of error and 1 if discarded
3931 */
3932
3933static int
3934htmlParseStartTag(htmlParserCtxtPtr ctxt) {
3935 const xmlChar *name;
3936 const xmlChar *attname;
3937 xmlChar *attvalue;
3938 const xmlChar **atts;
3939 int nbatts = 0;
3940 int maxatts;
3941 int meta = 0;
3942 int i;
3943 int discardtag = 0;
3944
3945 if ((ctxt == NULL) || (ctxt->input == NULL)) {
3946 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
3947 "htmlParseStartTag: context error\n", NULL, NULL);
3948 return -1;
3949 }
3950 if (ctxt->instate == XML_PARSER_EOF)
3951 return(-1);
3952 if (CUR != '<') return -1;
3953 NEXT;
3954
3955 atts = ctxt->atts;
3956 maxatts = ctxt->maxatts;
3957
3958 GROW;
3959 name = htmlParseHTMLName(ctxt);
3960 if (name == NULL) {
3961 htmlParseErr(ctxt, XML_ERR_NAME_REQUIRED,
3962 "htmlParseStartTag: invalid element name\n",
3963 NULL, NULL);
3964 /* Dump the bogus tag like browsers do */
3965 while ((CUR != 0) && (CUR != '>') &&
3966 (ctxt->instate != XML_PARSER_EOF))
3967 NEXT;
3968 return -1;
3969 }
3970 if (xmlStrEqual(name, BAD_CAST"meta"))
3971 meta = 1;
3972
3973 /*
3974 * Check for auto-closure of HTML elements.
3975 */
3976 htmlAutoClose(ctxt, name);
3977
3978 /*
3979 * Check for implied HTML elements.
3980 */
3981 htmlCheckImplied(ctxt, name);
3982
3983 /*
3984 * Avoid html at any level > 0, head at any level != 1
3985 * or any attempt to recurse body
3986 */
3987 if ((ctxt->nameNr > 0) && (xmlStrEqual(name, BAD_CAST"html"))) {
3988 htmlParseErr(ctxt, XML_HTML_STRUCURE_ERROR,
3989 "htmlParseStartTag: misplaced <html> tag\n",
3990 name, NULL);
3991 discardtag = 1;
3992 ctxt->depth++;
3993 }
3994 if ((ctxt->nameNr != 1) &&
3995 (xmlStrEqual(name, BAD_CAST"head"))) {
3996 htmlParseErr(ctxt, XML_HTML_STRUCURE_ERROR,
3997 "htmlParseStartTag: misplaced <head> tag\n",
3998 name, NULL);
3999 discardtag = 1;
4000 ctxt->depth++;
4001 }
4002 if (xmlStrEqual(name, BAD_CAST"body")) {
4003 int indx;
4004 for (indx = 0;indx < ctxt->nameNr;indx++) {
4005 if (xmlStrEqual(ctxt->nameTab[indx], BAD_CAST"body")) {
4006 htmlParseErr(ctxt, XML_HTML_STRUCURE_ERROR,
4007 "htmlParseStartTag: misplaced <body> tag\n",
4008 name, NULL);
4009 discardtag = 1;
4010 ctxt->depth++;
4011 }
4012 }
4013 }
4014
4015 /*
4016 * Now parse the attributes, it ends up with the ending
4017 *
4018 * (S Attribute)* S?
4019 */
4020 SKIP_BLANKS;
4021 while ((CUR != 0) &&
4022 (CUR != '>') &&
4023 ((CUR != '/') || (NXT(1) != '>'))) {
4024 GROW;
4025 attname = htmlParseAttribute(ctxt, &attvalue);
4026 if (attname != NULL) {
4027
4028 /*
4029 * Well formedness requires at most one declaration of an attribute
4030 */
4031 for (i = 0; i < nbatts;i += 2) {
4032 if (xmlStrEqual(atts[i], attname)) {
4033 htmlParseErr(ctxt, XML_ERR_ATTRIBUTE_REDEFINED,
4034 "Attribute %s redefined\n", attname, NULL);
4035 if (attvalue != NULL)
4036 xmlFree(attvalue);
4037 goto failed;
4038 }
4039 }
4040
4041 /*
4042 * Add the pair to atts
4043 */
4044 if (atts == NULL) {
4045 maxatts = 22; /* allow for 10 attrs by default */
4046 atts = (const xmlChar **)
4047 xmlMalloc(maxatts * sizeof(xmlChar *));
4048 if (atts == NULL) {
4049 htmlErrMemory(ctxt, NULL);
4050 if (attvalue != NULL)
4051 xmlFree(attvalue);
4052 goto failed;
4053 }
4054 ctxt->atts = atts;
4055 ctxt->maxatts = maxatts;
4056 } else if (nbatts + 4 > maxatts) {
4057 const xmlChar **n;
4058
4059 maxatts *= 2;
4060 n = (const xmlChar **) xmlRealloc((void *) atts,
4061 maxatts * sizeof(const xmlChar *));
4062 if (n == NULL) {
4063 htmlErrMemory(ctxt, NULL);
4064 if (attvalue != NULL)
4065 xmlFree(attvalue);
4066 goto failed;
4067 }
4068 atts = n;
4069 ctxt->atts = atts;
4070 ctxt->maxatts = maxatts;
4071 }
4072 atts[nbatts++] = attname;
4073 atts[nbatts++] = attvalue;
4074 atts[nbatts] = NULL;
4075 atts[nbatts + 1] = NULL;
4076 }
4077 else {
4078 if (attvalue != NULL)
4079 xmlFree(attvalue);
4080 /* Dump the bogus attribute string up to the next blank or
4081 * the end of the tag. */
4082 while ((CUR != 0) &&
4083 !(IS_BLANK_CH(CUR)) && (CUR != '>') &&
4084 ((CUR != '/') || (NXT(1) != '>')))
4085 NEXT;
4086 }
4087
4088failed:
4089 SKIP_BLANKS;
4090 }
4091
4092 /*
4093 * Handle specific association to the META tag
4094 */
4095 if (meta && (nbatts != 0))
4096 htmlCheckMeta(ctxt, atts);
4097
4098 /*
4099 * SAX: Start of Element !
4100 */
4101 if (!discardtag) {
4102 htmlnamePush(ctxt, name);
4103 if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL)) {
4104 if (nbatts != 0)
4105 ctxt->sax->startElement(ctxt->userData, name, atts);
4106 else
4107 ctxt->sax->startElement(ctxt->userData, name, NULL);
4108 }
4109 }
4110
4111 if (atts != NULL) {
4112 for (i = 1;i < nbatts;i += 2) {
4113 if (atts[i] != NULL)
4114 xmlFree((xmlChar *) atts[i]);
4115 }
4116 }
4117
4118 return(discardtag);
4119}
4120
4121/**
4122 * htmlParseEndTag:
4123 * @ctxt: an HTML parser context
4124 *
4125 * parse an end of tag
4126 *
4127 * [42] ETag ::= '</' Name S? '>'
4128 *
4129 * With namespace
4130 *
4131 * [NS 9] ETag ::= '</' QName S? '>'
4132 *
4133 * Returns 1 if the current level should be closed.
4134 */
4135
4136static int
4137htmlParseEndTag(htmlParserCtxtPtr ctxt)
4138{
4139 const xmlChar *name;
4140 const xmlChar *oldname;
4141 int i, ret;
4142
4143 if ((CUR != '<') || (NXT(1) != '/')) {
4144 htmlParseErr(ctxt, XML_ERR_LTSLASH_REQUIRED,
4145 "htmlParseEndTag: '</' not found\n", NULL, NULL);
4146 return (0);
4147 }
4148 SKIP(2);
4149
4150 name = htmlParseHTMLName(ctxt);
4151 if (name == NULL)
4152 return (0);
4153 /*
4154 * We should definitely be at the ending "S? '>'" part
4155 */
4156 SKIP_BLANKS;
4157 if (CUR != '>') {
4158 htmlParseErr(ctxt, XML_ERR_GT_REQUIRED,
4159 "End tag : expected '>'\n", NULL, NULL);
4160 /* Skip to next '>' */
4161 while ((CUR != 0) && (CUR != '>'))
4162 NEXT;
4163 }
4164 if (CUR == '>')
4165 NEXT;
4166
4167 /*
4168 * if we ignored misplaced tags in htmlParseStartTag don't pop them
4169 * out now.
4170 */
4171 if ((ctxt->depth > 0) &&
4172 (xmlStrEqual(name, BAD_CAST "html") ||
4173 xmlStrEqual(name, BAD_CAST "body") ||
4174 xmlStrEqual(name, BAD_CAST "head"))) {
4175 ctxt->depth--;
4176 return (0);
4177 }
4178
4179 /*
4180 * If the name read is not one of the element in the parsing stack
4181 * then return, it's just an error.
4182 */
4183 for (i = (ctxt->nameNr - 1); i >= 0; i--) {
4184 if (xmlStrEqual(name, ctxt->nameTab[i]))
4185 break;
4186 }
4187 if (i < 0) {
4188 htmlParseErr(ctxt, XML_ERR_TAG_NAME_MISMATCH,
4189 "Unexpected end tag : %s\n", name, NULL);
4190 return (0);
4191 }
4192
4193
4194 /*
4195 * Check for auto-closure of HTML elements.
4196 */
4197
4198 htmlAutoCloseOnClose(ctxt, name);
4199
4200 /*
4201 * Well formedness constraints, opening and closing must match.
4202 * With the exception that the autoclose may have popped stuff out
4203 * of the stack.
4204 */
4205 if ((ctxt->name != NULL) && (!xmlStrEqual(ctxt->name, name))) {
4206 htmlParseErr(ctxt, XML_ERR_TAG_NAME_MISMATCH,
4207 "Opening and ending tag mismatch: %s and %s\n",
4208 name, ctxt->name);
4209 }
4210
4211 /*
4212 * SAX: End of Tag
4213 */
4214 oldname = ctxt->name;
4215 if ((oldname != NULL) && (xmlStrEqual(oldname, name))) {
4216 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
4217 ctxt->sax->endElement(ctxt->userData, name);
4218 htmlNodeInfoPop(ctxt);
4219 htmlnamePop(ctxt);
4220 ret = 1;
4221 } else {
4222 ret = 0;
4223 }
4224
4225 return (ret);
4226}
4227
4228
4229/**
4230 * htmlParseReference:
4231 * @ctxt: an HTML parser context
4232 *
4233 * parse and handle entity references in content,
4234 * this will end-up in a call to character() since this is either a
4235 * CharRef, or a predefined entity.
4236 */
4237static void
4238htmlParseReference(htmlParserCtxtPtr ctxt) {
4239 const htmlEntityDesc * ent;
4240 xmlChar out[6];
4241 const xmlChar *name;
4242 if (CUR != '&') return;
4243
4244 if (NXT(1) == '#') {
4245 unsigned int c;
4246 int bits, i = 0;
4247
4248 c = htmlParseCharRef(ctxt);
4249 if (c == 0)
4250 return;
4251
4252 if (c < 0x80) { out[i++]= c; bits= -6; }
4253 else if (c < 0x800) { out[i++]=((c >> 6) & 0x1F) | 0xC0; bits= 0; }
4254 else if (c < 0x10000) { out[i++]=((c >> 12) & 0x0F) | 0xE0; bits= 6; }
4255 else { out[i++]=((c >> 18) & 0x07) | 0xF0; bits= 12; }
4256
4257 for ( ; bits >= 0; bits-= 6) {
4258 out[i++]= ((c >> bits) & 0x3F) | 0x80;
4259 }
4260 out[i] = 0;
4261
4262 htmlCheckParagraph(ctxt);
4263 if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
4264 ctxt->sax->characters(ctxt->userData, out, i);
4265 } else {
4266 ent = htmlParseEntityRef(ctxt, &name);
4267 if (name == NULL) {
4268 htmlCheckParagraph(ctxt);
4269 if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
4270 ctxt->sax->characters(ctxt->userData, BAD_CAST "&", 1);
4271 return;
4272 }
4273 if ((ent == NULL) || !(ent->value > 0)) {
4274 htmlCheckParagraph(ctxt);
4275 if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL)) {
4276 ctxt->sax->characters(ctxt->userData, BAD_CAST "&", 1);
4277 ctxt->sax->characters(ctxt->userData, name, xmlStrlen(name));
4278 /* ctxt->sax->characters(ctxt->userData, BAD_CAST ";", 1); */
4279 }
4280 } else {
4281 unsigned int c;
4282 int bits, i = 0;
4283
4284 c = ent->value;
4285 if (c < 0x80)
4286 { out[i++]= c; bits= -6; }
4287 else if (c < 0x800)
4288 { out[i++]=((c >> 6) & 0x1F) | 0xC0; bits= 0; }
4289 else if (c < 0x10000)
4290 { out[i++]=((c >> 12) & 0x0F) | 0xE0; bits= 6; }
4291 else
4292 { out[i++]=((c >> 18) & 0x07) | 0xF0; bits= 12; }
4293
4294 for ( ; bits >= 0; bits-= 6) {
4295 out[i++]= ((c >> bits) & 0x3F) | 0x80;
4296 }
4297 out[i] = 0;
4298
4299 htmlCheckParagraph(ctxt);
4300 if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
4301 ctxt->sax->characters(ctxt->userData, out, i);
4302 }
4303 }
4304}
4305
4306/**
4307 * htmlParseContent:
4308 * @ctxt: an HTML parser context
4309 *
4310 * Parse a content: comment, sub-element, reference or text.
4311 * Kept for compatibility with old code
4312 */
4313
4314static void
4315htmlParseContent(htmlParserCtxtPtr ctxt) {
4316 xmlChar *currentNode;
4317 int depth;
4318 const xmlChar *name;
4319
4320 currentNode = xmlStrdup(ctxt->name);
4321 depth = ctxt->nameNr;
4322 while (1) {
4323 GROW;
4324
4325 if (ctxt->instate == XML_PARSER_EOF)
4326 break;
4327
4328 /*
4329 * Our tag or one of it's parent or children is ending.
4330 */
4331 if ((CUR == '<') && (NXT(1) == '/')) {
4332 if (htmlParseEndTag(ctxt) &&
4333 ((currentNode != NULL) || (ctxt->nameNr == 0))) {
4334 if (currentNode != NULL)
4335 xmlFree(currentNode);
4336 return;
4337 }
4338 continue; /* while */
4339 }
4340
4341 else if ((CUR == '<') &&
4342 ((IS_ASCII_LETTER(NXT(1))) ||
4343 (NXT(1) == '_') || (NXT(1) == ':'))) {
4344 name = htmlParseHTMLName_nonInvasive(ctxt);
4345 if (name == NULL) {
4346 htmlParseErr(ctxt, XML_ERR_NAME_REQUIRED,
4347 "htmlParseStartTag: invalid element name\n",
4348 NULL, NULL);
4349 /* Dump the bogus tag like browsers do */
4350 while ((CUR != 0) && (CUR != '>'))
4351 NEXT;
4352
4353 if (currentNode != NULL)
4354 xmlFree(currentNode);
4355 return;
4356 }
4357
4358 if (ctxt->name != NULL) {
4359 if (htmlCheckAutoClose(name, ctxt->name) == 1) {
4360 htmlAutoClose(ctxt, name);
4361 continue;
4362 }
4363 }
4364 }
4365
4366 /*
4367 * Has this node been popped out during parsing of
4368 * the next element
4369 */
4370 if ((ctxt->nameNr > 0) && (depth >= ctxt->nameNr) &&
4371 (!xmlStrEqual(currentNode, ctxt->name)))
4372 {
4373 if (currentNode != NULL) xmlFree(currentNode);
4374 return;
4375 }
4376
4377 if ((CUR != 0) && ((xmlStrEqual(currentNode, BAD_CAST"script")) ||
4378 (xmlStrEqual(currentNode, BAD_CAST"style")))) {
4379 /*
4380 * Handle SCRIPT/STYLE separately
4381 */
4382 htmlParseScript(ctxt);
4383 } else {
4384 /*
4385 * Sometimes DOCTYPE arrives in the middle of the document
4386 */
4387 if ((CUR == '<') && (NXT(1) == '!') &&
4388 (UPP(2) == 'D') && (UPP(3) == 'O') &&
4389 (UPP(4) == 'C') && (UPP(5) == 'T') &&
4390 (UPP(6) == 'Y') && (UPP(7) == 'P') &&
4391 (UPP(8) == 'E')) {
4392 htmlParseErr(ctxt, XML_HTML_STRUCURE_ERROR,
4393 "Misplaced DOCTYPE declaration\n",
4394 BAD_CAST "DOCTYPE" , NULL);
4395 htmlParseDocTypeDecl(ctxt);
4396 }
4397
4398 /*
4399 * First case : a comment
4400 */
4401 if ((CUR == '<') && (NXT(1) == '!') &&
4402 (NXT(2) == '-') && (NXT(3) == '-')) {
4403 htmlParseComment(ctxt);
4404 }
4405
4406 /*
4407 * Second case : a Processing Instruction.
4408 */
4409 else if ((CUR == '<') && (NXT(1) == '?')) {
4410 htmlParsePI(ctxt);
4411 }
4412
4413 /*
4414 * Third case : a sub-element.
4415 */
4416 else if ((CUR == '<') && IS_ASCII_LETTER(NXT(1))) {
4417 htmlParseElement(ctxt);
4418 }
4419 else if (CUR == '<') {
4420 if ((ctxt->sax != NULL) && (!ctxt->disableSAX) &&
4421 (ctxt->sax->characters != NULL))
4422 ctxt->sax->characters(ctxt->userData, BAD_CAST "<", 1);
4423 NEXT;
4424 }
4425
4426 /*
4427 * Fourth case : a reference. If if has not been resolved,
4428 * parsing returns it's Name, create the node
4429 */
4430 else if (CUR == '&') {
4431 htmlParseReference(ctxt);
4432 }
4433
4434 /*
4435 * Fifth case : end of the resource
4436 */
4437 else if (CUR == 0) {
4438 htmlAutoCloseOnEnd(ctxt);
4439 break;
4440 }
4441
4442 /*
4443 * Last case, text. Note that References are handled directly.
4444 */
4445 else {
4446 htmlParseCharData(ctxt);
4447 }
4448 }
4449 GROW;
4450 }
4451 if (currentNode != NULL) xmlFree(currentNode);
4452}
4453
4454/**
4455 * htmlParseElement:
4456 * @ctxt: an HTML parser context
4457 *
4458 * parse an HTML element, this is highly recursive
4459 * this is kept for compatibility with previous code versions
4460 *
4461 * [39] element ::= EmptyElemTag | STag content ETag
4462 *
4463 * [41] Attribute ::= Name Eq AttValue
4464 */
4465
4466void
4467htmlParseElement(htmlParserCtxtPtr ctxt) {
4468 const xmlChar *name;
4469 xmlChar *currentNode = NULL;
4470 const htmlElemDesc * info;
4471 htmlParserNodeInfo node_info;
4472 int failed;
4473 int depth;
4474 const xmlChar *oldptr;
4475
4476 if ((ctxt == NULL) || (ctxt->input == NULL)) {
4477 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
4478 "htmlParseElement: context error\n", NULL, NULL);
4479 return;
4480 }
4481
4482 if (ctxt->instate == XML_PARSER_EOF)
4483 return;
4484
4485 /* Capture start position */
4486 if (ctxt->record_info) {
4487 node_info.begin_pos = ctxt->input->consumed +
4488 (CUR_PTR - ctxt->input->base);
4489 node_info.begin_line = ctxt->input->line;
4490 }
4491
4492 failed = htmlParseStartTag(ctxt);
4493 name = ctxt->name;
4494 if ((failed == -1) || (name == NULL)) {
4495 if (CUR == '>')
4496 NEXT;
4497 return;
4498 }
4499
4500 /*
4501 * Lookup the info for that element.
4502 */
4503 info = htmlTagLookup(name);
4504 if (info == NULL) {
4505 htmlParseErr(ctxt, XML_HTML_UNKNOWN_TAG,
4506 "Tag %s invalid\n", name, NULL);
4507 }
4508
4509 /*
4510 * Check for an Empty Element labeled the XML/SGML way
4511 */
4512 if ((CUR == '/') && (NXT(1) == '>')) {
4513 SKIP(2);
4514 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
4515 ctxt->sax->endElement(ctxt->userData, name);
4516 htmlnamePop(ctxt);
4517 return;
4518 }
4519
4520 if (CUR == '>') {
4521 NEXT;
4522 } else {
4523 htmlParseErr(ctxt, XML_ERR_GT_REQUIRED,
4524 "Couldn't find end of Start Tag %s\n", name, NULL);
4525
4526 /*
4527 * end of parsing of this node.
4528 */
4529 if (xmlStrEqual(name, ctxt->name)) {
4530 nodePop(ctxt);
4531 htmlnamePop(ctxt);
4532 }
4533
4534 /*
4535 * Capture end position and add node
4536 */
4537 if (ctxt->record_info) {
4538 node_info.end_pos = ctxt->input->consumed +
4539 (CUR_PTR - ctxt->input->base);
4540 node_info.end_line = ctxt->input->line;
4541 node_info.node = ctxt->node;
4542 xmlParserAddNodeInfo(ctxt, &node_info);
4543 }
4544 return;
4545 }
4546
4547 /*
4548 * Check for an Empty Element from DTD definition
4549 */
4550 if ((info != NULL) && (info->empty)) {
4551 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
4552 ctxt->sax->endElement(ctxt->userData, name);
4553 htmlnamePop(ctxt);
4554 return;
4555 }
4556
4557 /*
4558 * Parse the content of the element:
4559 */
4560 currentNode = xmlStrdup(ctxt->name);
4561 depth = ctxt->nameNr;
4562 while (CUR != 0) {
4563 oldptr = ctxt->input->cur;
4564 htmlParseContent(ctxt);
4565 if (oldptr==ctxt->input->cur) break;
4566 if (ctxt->nameNr < depth) break;
4567 }
4568
4569 /*
4570 * Capture end position and add node
4571 */
4572 if ( currentNode != NULL && ctxt->record_info ) {
4573 node_info.end_pos = ctxt->input->consumed +
4574 (CUR_PTR - ctxt->input->base);
4575 node_info.end_line = ctxt->input->line;
4576 node_info.node = ctxt->node;
4577 xmlParserAddNodeInfo(ctxt, &node_info);
4578 }
4579 if (CUR == 0) {
4580 htmlAutoCloseOnEnd(ctxt);
4581 }
4582
4583 if (currentNode != NULL)
4584 xmlFree(currentNode);
4585}
4586
4587static void
4588htmlParserFinishElementParsing(htmlParserCtxtPtr ctxt) {
4589 /*
4590 * Capture end position and add node
4591 */
4592 if ( ctxt->node != NULL && ctxt->record_info ) {
4593 ctxt->nodeInfo->end_pos = ctxt->input->consumed +
4594 (CUR_PTR - ctxt->input->base);
4595 ctxt->nodeInfo->end_line = ctxt->input->line;
4596 ctxt->nodeInfo->node = ctxt->node;
4597 xmlParserAddNodeInfo(ctxt, ctxt->nodeInfo);
4598 htmlNodeInfoPop(ctxt);
4599 }
4600 if (CUR == 0) {
4601 htmlAutoCloseOnEnd(ctxt);
4602 }
4603}
4604
4605/**
4606 * htmlParseElementInternal:
4607 * @ctxt: an HTML parser context
4608 *
4609 * parse an HTML element, new version, non recursive
4610 *
4611 * [39] element ::= EmptyElemTag | STag content ETag
4612 *
4613 * [41] Attribute ::= Name Eq AttValue
4614 */
4615
4616static void
4617htmlParseElementInternal(htmlParserCtxtPtr ctxt) {
4618 const xmlChar *name;
4619 const htmlElemDesc * info;
4620 htmlParserNodeInfo node_info = { NULL, 0, 0, 0, 0 };
4621 int failed;
4622
4623 if ((ctxt == NULL) || (ctxt->input == NULL)) {
4624 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
4625 "htmlParseElementInternal: context error\n", NULL, NULL);
4626 return;
4627 }
4628
4629 if (ctxt->instate == XML_PARSER_EOF)
4630 return;
4631
4632 /* Capture start position */
4633 if (ctxt->record_info) {
4634 node_info.begin_pos = ctxt->input->consumed +
4635 (CUR_PTR - ctxt->input->base);
4636 node_info.begin_line = ctxt->input->line;
4637 }
4638
4639 failed = htmlParseStartTag(ctxt);
4640 name = ctxt->name;
4641 if ((failed == -1) || (name == NULL)) {
4642 if (CUR == '>')
4643 NEXT;
4644 return;
4645 }
4646
4647 /*
4648 * Lookup the info for that element.
4649 */
4650 info = htmlTagLookup(name);
4651 if (info == NULL) {
4652 htmlParseErr(ctxt, XML_HTML_UNKNOWN_TAG,
4653 "Tag %s invalid\n", name, NULL);
4654 }
4655
4656 /*
4657 * Check for an Empty Element labeled the XML/SGML way
4658 */
4659 if ((CUR == '/') && (NXT(1) == '>')) {
4660 SKIP(2);
4661 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
4662 ctxt->sax->endElement(ctxt->userData, name);
4663 htmlnamePop(ctxt);
4664 return;
4665 }
4666
4667 if (CUR == '>') {
4668 NEXT;
4669 } else {
4670 htmlParseErr(ctxt, XML_ERR_GT_REQUIRED,
4671 "Couldn't find end of Start Tag %s\n", name, NULL);
4672
4673 /*
4674 * end of parsing of this node.
4675 */
4676 if (xmlStrEqual(name, ctxt->name)) {
4677 nodePop(ctxt);
4678 htmlnamePop(ctxt);
4679 }
4680
4681 if (ctxt->record_info)
4682 htmlNodeInfoPush(ctxt, &node_info);
4683 htmlParserFinishElementParsing(ctxt);
4684 return;
4685 }
4686
4687 /*
4688 * Check for an Empty Element from DTD definition
4689 */
4690 if ((info != NULL) && (info->empty)) {
4691 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
4692 ctxt->sax->endElement(ctxt->userData, name);
4693 htmlnamePop(ctxt);
4694 return;
4695 }
4696
4697 if (ctxt->record_info)
4698 htmlNodeInfoPush(ctxt, &node_info);
4699}
4700
4701/**
4702 * htmlParseContentInternal:
4703 * @ctxt: an HTML parser context
4704 *
4705 * Parse a content: comment, sub-element, reference or text.
4706 * New version for non recursive htmlParseElementInternal
4707 */
4708
4709static void
4710htmlParseContentInternal(htmlParserCtxtPtr ctxt) {
4711 xmlChar *currentNode;
4712 int depth;
4713 const xmlChar *name;
4714
4715 currentNode = xmlStrdup(ctxt->name);
4716 depth = ctxt->nameNr;
4717 while (1) {
4718 GROW;
4719
4720 if (ctxt->instate == XML_PARSER_EOF)
4721 break;
4722
4723 /*
4724 * Our tag or one of it's parent or children is ending.
4725 */
4726 if ((CUR == '<') && (NXT(1) == '/')) {
4727 if (htmlParseEndTag(ctxt) &&
4728 ((currentNode != NULL) || (ctxt->nameNr == 0))) {
4729 if (currentNode != NULL)
4730 xmlFree(currentNode);
4731
4732 currentNode = xmlStrdup(ctxt->name);
4733 depth = ctxt->nameNr;
4734 }
4735 continue; /* while */
4736 }
4737
4738 else if ((CUR == '<') &&
4739 ((IS_ASCII_LETTER(NXT(1))) ||
4740 (NXT(1) == '_') || (NXT(1) == ':'))) {
4741 name = htmlParseHTMLName_nonInvasive(ctxt);
4742 if (name == NULL) {
4743 htmlParseErr(ctxt, XML_ERR_NAME_REQUIRED,
4744 "htmlParseStartTag: invalid element name\n",
4745 NULL, NULL);
4746 /* Dump the bogus tag like browsers do */
4747 while ((CUR == 0) && (CUR != '>'))
4748 NEXT;
4749
4750 htmlParserFinishElementParsing(ctxt);
4751 if (currentNode != NULL)
4752 xmlFree(currentNode);
4753
4754 currentNode = xmlStrdup(ctxt->name);
4755 depth = ctxt->nameNr;
4756 continue;
4757 }
4758
4759 if (ctxt->name != NULL) {
4760 if (htmlCheckAutoClose(name, ctxt->name) == 1) {
4761 htmlAutoClose(ctxt, name);
4762 continue;
4763 }
4764 }
4765 }
4766
4767 /*
4768 * Has this node been popped out during parsing of
4769 * the next element
4770 */
4771 if ((ctxt->nameNr > 0) && (depth >= ctxt->nameNr) &&
4772 (!xmlStrEqual(currentNode, ctxt->name)))
4773 {
4774 htmlParserFinishElementParsing(ctxt);
4775 if (currentNode != NULL) xmlFree(currentNode);
4776
4777 currentNode = xmlStrdup(ctxt->name);
4778 depth = ctxt->nameNr;
4779 continue;
4780 }
4781
4782 if ((CUR != 0) && ((xmlStrEqual(currentNode, BAD_CAST"script")) ||
4783 (xmlStrEqual(currentNode, BAD_CAST"style")))) {
4784 /*
4785 * Handle SCRIPT/STYLE separately
4786 */
4787 htmlParseScript(ctxt);
4788 } else {
4789 /*
4790 * Sometimes DOCTYPE arrives in the middle of the document
4791 */
4792 if ((CUR == '<') && (NXT(1) == '!') &&
4793 (UPP(2) == 'D') && (UPP(3) == 'O') &&
4794 (UPP(4) == 'C') && (UPP(5) == 'T') &&
4795 (UPP(6) == 'Y') && (UPP(7) == 'P') &&
4796 (UPP(8) == 'E')) {
4797 htmlParseErr(ctxt, XML_HTML_STRUCURE_ERROR,
4798 "Misplaced DOCTYPE declaration\n",
4799 BAD_CAST "DOCTYPE" , NULL);
4800 htmlParseDocTypeDecl(ctxt);
4801 }
4802
4803 /*
4804 * First case : a comment
4805 */
4806 if ((CUR == '<') && (NXT(1) == '!') &&
4807 (NXT(2) == '-') && (NXT(3) == '-')) {
4808 htmlParseComment(ctxt);
4809 }
4810
4811 /*
4812 * Second case : a Processing Instruction.
4813 */
4814 else if ((CUR == '<') && (NXT(1) == '?')) {
4815 htmlParsePI(ctxt);
4816 }
4817
4818 /*
4819 * Third case : a sub-element.
4820 */
4821 else if ((CUR == '<') && IS_ASCII_LETTER(NXT(1))) {
4822 htmlParseElementInternal(ctxt);
4823 if (currentNode != NULL) xmlFree(currentNode);
4824
4825 currentNode = xmlStrdup(ctxt->name);
4826 depth = ctxt->nameNr;
4827 }
4828 else if (CUR == '<') {
4829 if ((ctxt->sax != NULL) && (!ctxt->disableSAX) &&
4830 (ctxt->sax->characters != NULL))
4831 ctxt->sax->characters(ctxt->userData, BAD_CAST "<", 1);
4832 NEXT;
4833 }
4834
4835 /*
4836 * Fourth case : a reference. If if has not been resolved,
4837 * parsing returns it's Name, create the node
4838 */
4839 else if (CUR == '&') {
4840 htmlParseReference(ctxt);
4841 }
4842
4843 /*
4844 * Fifth case : end of the resource
4845 */
4846 else if (CUR == 0) {
4847 htmlAutoCloseOnEnd(ctxt);
4848 break;
4849 }
4850
4851 /*
4852 * Last case, text. Note that References are handled directly.
4853 */
4854 else {
4855 htmlParseCharData(ctxt);
4856 }
4857 }
4858 GROW;
4859 }
4860 if (currentNode != NULL) xmlFree(currentNode);
4861}
4862
4863/**
4864 * htmlParseContent:
4865 * @ctxt: an HTML parser context
4866 *
4867 * Parse a content: comment, sub-element, reference or text.
4868 * This is the entry point when called from parser.c
4869 */
4870
4871void
4872__htmlParseContent(void *ctxt) {
4873 if (ctxt != NULL)
4874 htmlParseContentInternal((htmlParserCtxtPtr) ctxt);
4875}
4876
4877/**
4878 * htmlParseDocument:
4879 * @ctxt: an HTML parser context
4880 *
4881 * parse an HTML document (and build a tree if using the standard SAX
4882 * interface).
4883 *
4884 * Returns 0, -1 in case of error. the parser context is augmented
4885 * as a result of the parsing.
4886 */
4887
4888int
4889htmlParseDocument(htmlParserCtxtPtr ctxt) {
4890 xmlChar start[4];
4891 xmlCharEncoding enc;
4892 xmlDtdPtr dtd;
4893
4894 xmlInitParser();
4895
4896 htmlDefaultSAXHandlerInit();
4897
4898 if ((ctxt == NULL) || (ctxt->input == NULL)) {
4899 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
4900 "htmlParseDocument: context error\n", NULL, NULL);
4901 return(XML_ERR_INTERNAL_ERROR);
4902 }
4903 ctxt->html = 1;
4904 ctxt->linenumbers = 1;
4905 GROW;
4906 /*
4907 * SAX: beginning of the document processing.
4908 */
4909 if ((ctxt->sax) && (ctxt->sax->setDocumentLocator))
4910 ctxt->sax->setDocumentLocator(ctxt->userData, &xmlDefaultSAXLocator);
4911
4912 if ((ctxt->encoding == (const xmlChar *)XML_CHAR_ENCODING_NONE) &&
4913 ((ctxt->input->end - ctxt->input->cur) >= 4)) {
4914 /*
4915 * Get the 4 first bytes and decode the charset
4916 * if enc != XML_CHAR_ENCODING_NONE
4917 * plug some encoding conversion routines.
4918 */
4919 start[0] = RAW;
4920 start[1] = NXT(1);
4921 start[2] = NXT(2);
4922 start[3] = NXT(3);
4923 enc = xmlDetectCharEncoding(&start[0], 4);
4924 if (enc != XML_CHAR_ENCODING_NONE) {
4925 xmlSwitchEncoding(ctxt, enc);
4926 }
4927 }
4928
4929 /*
4930 * Wipe out everything which is before the first '<'
4931 */
4932 SKIP_BLANKS;
4933 if (CUR == 0) {
4934 htmlParseErr(ctxt, XML_ERR_DOCUMENT_EMPTY,
4935 "Document is empty\n", NULL, NULL);
4936 }
4937
4938 if ((ctxt->sax) && (ctxt->sax->startDocument) && (!ctxt->disableSAX))
4939 ctxt->sax->startDocument(ctxt->userData);
4940
4941
4942 /*
4943 * Parse possible comments and PIs before any content
4944 */
4945 while (((CUR == '<') && (NXT(1) == '!') &&
4946 (NXT(2) == '-') && (NXT(3) == '-')) ||
4947 ((CUR == '<') && (NXT(1) == '?'))) {
4948 htmlParseComment(ctxt);
4949 htmlParsePI(ctxt);
4950 SKIP_BLANKS;
4951 }
4952
4953
4954 /*
4955 * Then possibly doc type declaration(s) and more Misc
4956 * (doctypedecl Misc*)?
4957 */
4958 if ((CUR == '<') && (NXT(1) == '!') &&
4959 (UPP(2) == 'D') && (UPP(3) == 'O') &&
4960 (UPP(4) == 'C') && (UPP(5) == 'T') &&
4961 (UPP(6) == 'Y') && (UPP(7) == 'P') &&
4962 (UPP(8) == 'E')) {
4963 htmlParseDocTypeDecl(ctxt);
4964 }
4965 SKIP_BLANKS;
4966
4967 /*
4968 * Parse possible comments and PIs before any content
4969 */
4970 while (((CUR == '<') && (NXT(1) == '!') &&
4971 (NXT(2) == '-') && (NXT(3) == '-')) ||
4972 ((CUR == '<') && (NXT(1) == '?'))) {
4973 htmlParseComment(ctxt);
4974 htmlParsePI(ctxt);
4975 SKIP_BLANKS;
4976 }
4977
4978 /*
4979 * Time to start parsing the tree itself
4980 */
4981 htmlParseContentInternal(ctxt);
4982
4983 /*
4984 * autoclose
4985 */
4986 if (CUR == 0)
4987 htmlAutoCloseOnEnd(ctxt);
4988
4989
4990 /*
4991 * SAX: end of the document processing.
4992 */
4993 if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
4994 ctxt->sax->endDocument(ctxt->userData);
4995
4996 if ((!(ctxt->options & HTML_PARSE_NODEFDTD)) && (ctxt->myDoc != NULL)) {
4997 dtd = xmlGetIntSubset(ctxt->myDoc);
4998 if (dtd == NULL)
4999 ctxt->myDoc->intSubset =
5000 xmlCreateIntSubset(ctxt->myDoc, BAD_CAST "html",
5001 BAD_CAST "-//W3C//DTD HTML 4.0 Transitional//EN",
5002 BAD_CAST "http://www.w3.org/TR/REC-html40/loose.dtd");
5003 }
5004 if (! ctxt->wellFormed) return(-1);
5005 return(0);
5006}
5007
5008
5009/************************************************************************
5010 * *
5011 * Parser contexts handling *
5012 * *
5013 ************************************************************************/
5014
5015/**
5016 * htmlInitParserCtxt:
5017 * @ctxt: an HTML parser context
5018 *
5019 * Initialize a parser context
5020 *
5021 * Returns 0 in case of success and -1 in case of error
5022 */
5023
5024static int
5025htmlInitParserCtxt(htmlParserCtxtPtr ctxt)
5026{
5027 htmlSAXHandler *sax;
5028
5029 if (ctxt == NULL) return(-1);
5030 memset(ctxt, 0, sizeof(htmlParserCtxt));
5031
5032 ctxt->dict = xmlDictCreate();
5033 if (ctxt->dict == NULL) {
5034 htmlErrMemory(NULL, "htmlInitParserCtxt: out of memory\n");
5035 return(-1);
5036 }
5037 sax = (htmlSAXHandler *) xmlMalloc(sizeof(htmlSAXHandler));
5038 if (sax == NULL) {
5039 htmlErrMemory(NULL, "htmlInitParserCtxt: out of memory\n");
5040 return(-1);
5041 }
5042 else
5043 memset(sax, 0, sizeof(htmlSAXHandler));
5044
5045 /* Allocate the Input stack */
5046 ctxt->inputTab = (htmlParserInputPtr *)
5047 xmlMalloc(5 * sizeof(htmlParserInputPtr));
5048 if (ctxt->inputTab == NULL) {
5049 htmlErrMemory(NULL, "htmlInitParserCtxt: out of memory\n");
5050 ctxt->inputNr = 0;
5051 ctxt->inputMax = 0;
5052 ctxt->input = NULL;
5053 return(-1);
5054 }
5055 ctxt->inputNr = 0;
5056 ctxt->inputMax = 5;
5057 ctxt->input = NULL;
5058 ctxt->version = NULL;
5059 ctxt->encoding = NULL;
5060 ctxt->standalone = -1;
5061 ctxt->instate = XML_PARSER_START;
5062
5063 /* Allocate the Node stack */
5064 ctxt->nodeTab = (htmlNodePtr *) xmlMalloc(10 * sizeof(htmlNodePtr));
5065 if (ctxt->nodeTab == NULL) {
5066 htmlErrMemory(NULL, "htmlInitParserCtxt: out of memory\n");
5067 ctxt->nodeNr = 0;
5068 ctxt->nodeMax = 0;
5069 ctxt->node = NULL;
5070 ctxt->inputNr = 0;
5071 ctxt->inputMax = 0;
5072 ctxt->input = NULL;
5073 return(-1);
5074 }
5075 ctxt->nodeNr = 0;
5076 ctxt->nodeMax = 10;
5077 ctxt->node = NULL;
5078
5079 /* Allocate the Name stack */
5080 ctxt->nameTab = (const xmlChar **) xmlMalloc(10 * sizeof(xmlChar *));
5081 if (ctxt->nameTab == NULL) {
5082 htmlErrMemory(NULL, "htmlInitParserCtxt: out of memory\n");
5083 ctxt->nameNr = 0;
5084 ctxt->nameMax = 0;
5085 ctxt->name = NULL;
5086 ctxt->nodeNr = 0;
5087 ctxt->nodeMax = 0;
5088 ctxt->node = NULL;
5089 ctxt->inputNr = 0;
5090 ctxt->inputMax = 0;
5091 ctxt->input = NULL;
5092 return(-1);
5093 }
5094 ctxt->nameNr = 0;
5095 ctxt->nameMax = 10;
5096 ctxt->name = NULL;
5097
5098 ctxt->nodeInfoTab = NULL;
5099 ctxt->nodeInfoNr = 0;
5100 ctxt->nodeInfoMax = 0;
5101
5102 if (sax == NULL) ctxt->sax = (xmlSAXHandlerPtr) &htmlDefaultSAXHandler;
5103 else {
5104 ctxt->sax = sax;
5105 memcpy(sax, &htmlDefaultSAXHandler, sizeof(xmlSAXHandlerV1));
5106 }
5107 ctxt->userData = ctxt;
5108 ctxt->myDoc = NULL;
5109 ctxt->wellFormed = 1;
5110 ctxt->replaceEntities = 0;
5111 ctxt->linenumbers = xmlLineNumbersDefaultValue;
5112 ctxt->keepBlanks = xmlKeepBlanksDefaultValue;
5113 ctxt->html = 1;
5114 ctxt->vctxt.finishDtd = XML_CTXT_FINISH_DTD_0;
5115 ctxt->vctxt.userData = ctxt;
5116 ctxt->vctxt.error = xmlParserValidityError;
5117 ctxt->vctxt.warning = xmlParserValidityWarning;
5118 ctxt->record_info = 0;
5119 ctxt->validate = 0;
5120 ctxt->checkIndex = 0;
5121 ctxt->catalogs = NULL;
5122 xmlInitNodeInfoSeq(&ctxt->node_seq);
5123 return(0);
5124}
5125
5126/**
5127 * htmlFreeParserCtxt:
5128 * @ctxt: an HTML parser context
5129 *
5130 * Free all the memory used by a parser context. However the parsed
5131 * document in ctxt->myDoc is not freed.
5132 */
5133
5134void
5135htmlFreeParserCtxt(htmlParserCtxtPtr ctxt)
5136{
5137 xmlFreeParserCtxt(ctxt);
5138}
5139
5140/**
5141 * htmlNewParserCtxt:
5142 *
5143 * Allocate and initialize a new parser context.
5144 *
5145 * Returns the htmlParserCtxtPtr or NULL in case of allocation error
5146 */
5147
5148htmlParserCtxtPtr
5149htmlNewParserCtxt(void)
5150{
5151 xmlParserCtxtPtr ctxt;
5152
5153 ctxt = (xmlParserCtxtPtr) xmlMalloc(sizeof(xmlParserCtxt));
5154 if (ctxt == NULL) {
5155 htmlErrMemory(NULL, "NewParserCtxt: out of memory\n");
5156 return(NULL);
5157 }
5158 memset(ctxt, 0, sizeof(xmlParserCtxt));
5159 if (htmlInitParserCtxt(ctxt) < 0) {
5160 htmlFreeParserCtxt(ctxt);
5161 return(NULL);
5162 }
5163 return(ctxt);
5164}
5165
5166/**
5167 * htmlCreateMemoryParserCtxt:
5168 * @buffer: a pointer to a char array
5169 * @size: the size of the array
5170 *
5171 * Create a parser context for an HTML in-memory document.
5172 *
5173 * Returns the new parser context or NULL
5174 */
5175htmlParserCtxtPtr
5176htmlCreateMemoryParserCtxt(const char *buffer, int size) {
5177 xmlParserCtxtPtr ctxt;
5178 xmlParserInputPtr input;
5179 xmlParserInputBufferPtr buf;
5180
5181 if (buffer == NULL)
5182 return(NULL);
5183 if (size <= 0)
5184 return(NULL);
5185
5186 ctxt = htmlNewParserCtxt();
5187 if (ctxt == NULL)
5188 return(NULL);
5189
5190 buf = xmlParserInputBufferCreateMem(buffer, size, XML_CHAR_ENCODING_NONE);
5191 if (buf == NULL) return(NULL);
5192
5193 input = xmlNewInputStream(ctxt);
5194 if (input == NULL) {
5195 xmlFreeParserInputBuffer(buf);
5196 xmlFreeParserCtxt(ctxt);
5197 return(NULL);
5198 }
5199
5200 input->filename = NULL;
5201 input->buf = buf;
5202 xmlBufResetInput(buf->buffer, input);
5203
5204 inputPush(ctxt, input);
5205 return(ctxt);
5206}
5207
5208/**
5209 * htmlCreateDocParserCtxt:
5210 * @cur: a pointer to an array of xmlChar
5211 * @encoding: a free form C string describing the HTML document encoding, or NULL
5212 *
5213 * Create a parser context for an HTML document.
5214 *
5215 * TODO: check the need to add encoding handling there
5216 *
5217 * Returns the new parser context or NULL
5218 */
5219static htmlParserCtxtPtr
5220htmlCreateDocParserCtxt(const xmlChar *cur, const char *encoding) {
5221 int len;
5222 htmlParserCtxtPtr ctxt;
5223
5224 if (cur == NULL)
5225 return(NULL);
5226 len = xmlStrlen(cur);
5227 ctxt = htmlCreateMemoryParserCtxt((char *)cur, len);
5228 if (ctxt == NULL)
5229 return(NULL);
5230
5231 if (encoding != NULL) {
5232 xmlCharEncoding enc;
5233 xmlCharEncodingHandlerPtr handler;
5234
5235 if (ctxt->input->encoding != NULL)
5236 xmlFree((xmlChar *) ctxt->input->encoding);
5237 ctxt->input->encoding = xmlStrdup((const xmlChar *) encoding);
5238
5239 enc = xmlParseCharEncoding(encoding);
5240 /*
5241 * registered set of known encodings
5242 */
5243 if (enc != XML_CHAR_ENCODING_ERROR) {
5244 xmlSwitchEncoding(ctxt, enc);
5245 if (ctxt->errNo == XML_ERR_UNSUPPORTED_ENCODING) {
5246 htmlParseErr(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
5247 "Unsupported encoding %s\n",
5248 (const xmlChar *) encoding, NULL);
5249 }
5250 } else {
5251 /*
5252 * fallback for unknown encodings
5253 */
5254 handler = xmlFindCharEncodingHandler((const char *) encoding);
5255 if (handler != NULL) {
5256 xmlSwitchToEncoding(ctxt, handler);
5257 } else {
5258 htmlParseErr(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
5259 "Unsupported encoding %s\n",
5260 (const xmlChar *) encoding, NULL);
5261 }
5262 }
5263 }
5264 return(ctxt);
5265}
5266
5267#ifdef LIBXML_PUSH_ENABLED
5268/************************************************************************
5269 * *
5270 * Progressive parsing interfaces *
5271 * *
5272 ************************************************************************/
5273
5274/**
5275 * htmlParseLookupSequence:
5276 * @ctxt: an HTML parser context
5277 * @first: the first char to lookup
5278 * @next: the next char to lookup or zero
5279 * @third: the next char to lookup or zero
5280 * @ignoreattrval: skip over attribute values
5281 *
5282 * Try to find if a sequence (first, next, third) or just (first next) or
5283 * (first) is available in the input stream.
5284 * This function has a side effect of (possibly) incrementing ctxt->checkIndex
5285 * to avoid rescanning sequences of bytes, it DOES change the state of the
5286 * parser, do not use liberally.
5287 * This is basically similar to xmlParseLookupSequence()
5288 *
5289 * Returns the index to the current parsing point if the full sequence
5290 * is available, -1 otherwise.
5291 */
5292static int
5293htmlParseLookupSequence(htmlParserCtxtPtr ctxt, xmlChar first,
5294 xmlChar next, xmlChar third, int ignoreattrval)
5295{
5296 int base, len;
5297 htmlParserInputPtr in;
5298 const xmlChar *buf;
5299 int invalue = 0;
5300 char valdellim = 0x0;
5301
5302 in = ctxt->input;
5303 if (in == NULL)
5304 return (-1);
5305
5306 base = in->cur - in->base;
5307 if (base < 0)
5308 return (-1);
5309
5310 if (ctxt->checkIndex > base) {
5311 base = ctxt->checkIndex;
5312 /* Abuse hasPErefs member to restore current state. */
5313 invalue = ctxt->hasPErefs & 1 ? 1 : 0;
5314 }
5315
5316 if (in->buf == NULL) {
5317 buf = in->base;
5318 len = in->length;
5319 } else {
5320 buf = xmlBufContent(in->buf->buffer);
5321 len = xmlBufUse(in->buf->buffer);
5322 }
5323
5324 /* take into account the sequence length */
5325 if (third)
5326 len -= 2;
5327 else if (next)
5328 len--;
5329 for (; base < len; base++) {
5330 if (ignoreattrval) {
5331 if (buf[base] == '"' || buf[base] == '\'') {
5332 if (invalue) {
5333 if (buf[base] == valdellim) {
5334 invalue = 0;
5335 continue;
5336 }
5337 } else {
5338 valdellim = buf[base];
5339 invalue = 1;
5340 continue;
5341 }
5342 } else if (invalue) {
5343 continue;
5344 }
5345 }
5346 if (buf[base] == first) {
5347 if (third != 0) {
5348 if ((buf[base + 1] != next) || (buf[base + 2] != third))
5349 continue;
5350 } else if (next != 0) {
5351 if (buf[base + 1] != next)
5352 continue;
5353 }
5354 ctxt->checkIndex = 0;
5355#ifdef DEBUG_PUSH
5356 if (next == 0)
5357 xmlGenericError(xmlGenericErrorContext,
5358 "HPP: lookup '%c' found at %d\n",
5359 first, base);
5360 else if (third == 0)
5361 xmlGenericError(xmlGenericErrorContext,
5362 "HPP: lookup '%c%c' found at %d\n",
5363 first, next, base);
5364 else
5365 xmlGenericError(xmlGenericErrorContext,
5366 "HPP: lookup '%c%c%c' found at %d\n",
5367 first, next, third, base);
5368#endif
5369 return (base - (in->cur - in->base));
5370 }
5371 }
5372 ctxt->checkIndex = base;
5373 /* Abuse hasPErefs member to track current state. */
5374 if (invalue)
5375 ctxt->hasPErefs |= 1;
5376 else
5377 ctxt->hasPErefs &= ~1;
5378#ifdef DEBUG_PUSH
5379 if (next == 0)
5380 xmlGenericError(xmlGenericErrorContext,
5381 "HPP: lookup '%c' failed\n", first);
5382 else if (third == 0)
5383 xmlGenericError(xmlGenericErrorContext,
5384 "HPP: lookup '%c%c' failed\n", first, next);
5385 else
5386 xmlGenericError(xmlGenericErrorContext,
5387 "HPP: lookup '%c%c%c' failed\n", first, next,
5388 third);
5389#endif
5390 return (-1);
5391}
5392
5393/**
5394 * htmlParseLookupCommentEnd:
5395 * @ctxt: an HTML parser context
5396 *
5397 * Try to find a comment end tag in the input stream
5398 * The search includes "-->" as well as WHATWG-recommended incorrectly-closed tags.
5399 * (See https://html.spec.whatwg.org/multipage/parsing.html#parse-error-incorrectly-closed-comment)
5400 * This function has a side effect of (possibly) incrementing ctxt->checkIndex
5401 * to avoid rescanning sequences of bytes, it DOES change the state of the
5402 * parser, do not use liberally.
5403 * This wraps to htmlParseLookupSequence()
5404 *
5405 * Returns the index to the current parsing point if the full sequence is available, -1 otherwise.
5406 */
5407static int
5408htmlParseLookupCommentEnd(htmlParserCtxtPtr ctxt)
5409{
5410 int mark = 0;
5411 int cur = CUR_PTR - BASE_PTR;
5412
5413 while (mark >= 0) {
5414 mark = htmlParseLookupSequence(ctxt, '-', '-', 0, 0);
5415 if ((mark < 0) ||
5416 (NXT(mark+2) == '>') ||
5417 ((NXT(mark+2) == '!') && (NXT(mark+3) == '>'))) {
5418 return mark;
5419 }
5420 ctxt->checkIndex = cur + mark + 1;
5421 }
5422 return mark;
5423}
5424
5425
5426/**
5427 * htmlParseTryOrFinish:
5428 * @ctxt: an HTML parser context
5429 * @terminate: last chunk indicator
5430 *
5431 * Try to progress on parsing
5432 *
5433 * Returns zero if no parsing was possible
5434 */
5435static int
5436htmlParseTryOrFinish(htmlParserCtxtPtr ctxt, int terminate) {
5437 int ret = 0;
5438 htmlParserInputPtr in;
5439 ptrdiff_t avail = 0;
5440 xmlChar cur, next;
5441
5442 htmlParserNodeInfo node_info;
5443
5444#ifdef DEBUG_PUSH
5445 switch (ctxt->instate) {
5446 case XML_PARSER_EOF:
5447 xmlGenericError(xmlGenericErrorContext,
5448 "HPP: try EOF\n"); break;
5449 case XML_PARSER_START:
5450 xmlGenericError(xmlGenericErrorContext,
5451 "HPP: try START\n"); break;
5452 case XML_PARSER_MISC:
5453 xmlGenericError(xmlGenericErrorContext,
5454 "HPP: try MISC\n");break;
5455 case XML_PARSER_COMMENT:
5456 xmlGenericError(xmlGenericErrorContext,
5457 "HPP: try COMMENT\n");break;
5458 case XML_PARSER_PROLOG:
5459 xmlGenericError(xmlGenericErrorContext,
5460 "HPP: try PROLOG\n");break;
5461 case XML_PARSER_START_TAG:
5462 xmlGenericError(xmlGenericErrorContext,
5463 "HPP: try START_TAG\n");break;
5464 case XML_PARSER_CONTENT:
5465 xmlGenericError(xmlGenericErrorContext,
5466 "HPP: try CONTENT\n");break;
5467 case XML_PARSER_CDATA_SECTION:
5468 xmlGenericError(xmlGenericErrorContext,
5469 "HPP: try CDATA_SECTION\n");break;
5470 case XML_PARSER_END_TAG:
5471 xmlGenericError(xmlGenericErrorContext,
5472 "HPP: try END_TAG\n");break;
5473 case XML_PARSER_ENTITY_DECL:
5474 xmlGenericError(xmlGenericErrorContext,
5475 "HPP: try ENTITY_DECL\n");break;
5476 case XML_PARSER_ENTITY_VALUE:
5477 xmlGenericError(xmlGenericErrorContext,
5478 "HPP: try ENTITY_VALUE\n");break;
5479 case XML_PARSER_ATTRIBUTE_VALUE:
5480 xmlGenericError(xmlGenericErrorContext,
5481 "HPP: try ATTRIBUTE_VALUE\n");break;
5482 case XML_PARSER_DTD:
5483 xmlGenericError(xmlGenericErrorContext,
5484 "HPP: try DTD\n");break;
5485 case XML_PARSER_EPILOG:
5486 xmlGenericError(xmlGenericErrorContext,
5487 "HPP: try EPILOG\n");break;
5488 case XML_PARSER_PI:
5489 xmlGenericError(xmlGenericErrorContext,
5490 "HPP: try PI\n");break;
5491 case XML_PARSER_SYSTEM_LITERAL:
5492 xmlGenericError(xmlGenericErrorContext,
5493 "HPP: try SYSTEM_LITERAL\n");break;
5494 }
5495#endif
5496
5497 while (1) {
5498
5499 in = ctxt->input;
5500 if (in == NULL) break;
5501 if (in->buf == NULL)
5502 avail = in->length - (in->cur - in->base);
5503 else
5504 avail = (ptrdiff_t)xmlBufUse(in->buf->buffer) -
5505 (in->cur - in->base);
5506 if ((avail == 0) && (terminate)) {
5507 htmlAutoCloseOnEnd(ctxt);
5508 if ((ctxt->nameNr == 0) && (ctxt->instate != XML_PARSER_EOF)) {
5509 /*
5510 * SAX: end of the document processing.
5511 */
5512 ctxt->instate = XML_PARSER_EOF;
5513 if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
5514 ctxt->sax->endDocument(ctxt->userData);
5515 }
5516 }
5517 if (avail < 1)
5518 goto done;
5519 /*
5520 * This is done to make progress and avoid an infinite loop
5521 * if a parsing attempt was aborted by hitting a NUL byte. After
5522 * changing htmlCurrentChar, this probably isn't necessary anymore.
5523 * We should consider removing this check.
5524 */
5525 cur = in->cur[0];
5526 if (cur == 0) {
5527 SKIP(1);
5528 continue;
5529 }
5530
5531 switch (ctxt->instate) {
5532 case XML_PARSER_EOF:
5533 /*
5534 * Document parsing is done !
5535 */
5536 goto done;
5537 case XML_PARSER_START:
5538 /*
5539 * Very first chars read from the document flow.
5540 */
5541 cur = in->cur[0];
5542 if (IS_BLANK_CH(cur)) {
5543 SKIP_BLANKS;
5544 if (in->buf == NULL)
5545 avail = in->length - (in->cur - in->base);
5546 else
5547 avail = (ptrdiff_t)xmlBufUse(in->buf->buffer) -
5548 (in->cur - in->base);
5549 }
5550 if ((ctxt->sax) && (ctxt->sax->setDocumentLocator))
5551 ctxt->sax->setDocumentLocator(ctxt->userData,
5552 &xmlDefaultSAXLocator);
5553 if ((ctxt->sax) && (ctxt->sax->startDocument) &&
5554 (!ctxt->disableSAX))
5555 ctxt->sax->startDocument(ctxt->userData);
5556
5557 cur = in->cur[0];
5558 next = in->cur[1];
5559 if ((cur == '<') && (next == '!') &&
5560 (UPP(2) == 'D') && (UPP(3) == 'O') &&
5561 (UPP(4) == 'C') && (UPP(5) == 'T') &&
5562 (UPP(6) == 'Y') && (UPP(7) == 'P') &&
5563 (UPP(8) == 'E')) {
5564 if ((!terminate) &&
5565 (htmlParseLookupSequence(ctxt, '>', 0, 0, 1) < 0))
5566 goto done;
5567#ifdef DEBUG_PUSH
5568 xmlGenericError(xmlGenericErrorContext,
5569 "HPP: Parsing internal subset\n");
5570#endif
5571 htmlParseDocTypeDecl(ctxt);
5572 ctxt->instate = XML_PARSER_PROLOG;
5573#ifdef DEBUG_PUSH
5574 xmlGenericError(xmlGenericErrorContext,
5575 "HPP: entering PROLOG\n");
5576#endif
5577 } else {
5578 ctxt->instate = XML_PARSER_MISC;
5579#ifdef DEBUG_PUSH
5580 xmlGenericError(xmlGenericErrorContext,
5581 "HPP: entering MISC\n");
5582#endif
5583 }
5584 break;
5585 case XML_PARSER_MISC:
5586 SKIP_BLANKS;
5587 if (in->buf == NULL)
5588 avail = in->length - (in->cur - in->base);
5589 else
5590 avail = (ptrdiff_t)xmlBufUse(in->buf->buffer) -
5591 (in->cur - in->base);
5592 /*
5593 * no chars in buffer
5594 */
5595 if (avail < 1)
5596 goto done;
5597 /*
5598 * not enough chars in buffer
5599 */
5600 if (avail < 2) {
5601 if (!terminate)
5602 goto done;
5603 else
5604 next = ' ';
5605 } else {
5606 next = in->cur[1];
5607 }
5608 cur = in->cur[0];
5609 if ((cur == '<') && (next == '!') &&
5610 (in->cur[2] == '-') && (in->cur[3] == '-')) {
5611 if ((!terminate) && (htmlParseLookupCommentEnd(ctxt) < 0))
5612 goto done;
5613#ifdef DEBUG_PUSH
5614 xmlGenericError(xmlGenericErrorContext,
5615 "HPP: Parsing Comment\n");
5616#endif
5617 htmlParseComment(ctxt);
5618 ctxt->instate = XML_PARSER_MISC;
5619 } else if ((cur == '<') && (next == '?')) {
5620 if ((!terminate) &&
5621 (htmlParseLookupSequence(ctxt, '>', 0, 0, 0) < 0))
5622 goto done;
5623#ifdef DEBUG_PUSH
5624 xmlGenericError(xmlGenericErrorContext,
5625 "HPP: Parsing PI\n");
5626#endif
5627 htmlParsePI(ctxt);
5628 ctxt->instate = XML_PARSER_MISC;
5629 } else if ((cur == '<') && (next == '!') &&
5630 (UPP(2) == 'D') && (UPP(3) == 'O') &&
5631 (UPP(4) == 'C') && (UPP(5) == 'T') &&
5632 (UPP(6) == 'Y') && (UPP(7) == 'P') &&
5633 (UPP(8) == 'E')) {
5634 if ((!terminate) &&
5635 (htmlParseLookupSequence(ctxt, '>', 0, 0, 1) < 0))
5636 goto done;
5637#ifdef DEBUG_PUSH
5638 xmlGenericError(xmlGenericErrorContext,
5639 "HPP: Parsing internal subset\n");
5640#endif
5641 htmlParseDocTypeDecl(ctxt);
5642 ctxt->instate = XML_PARSER_PROLOG;
5643#ifdef DEBUG_PUSH
5644 xmlGenericError(xmlGenericErrorContext,
5645 "HPP: entering PROLOG\n");
5646#endif
5647 } else if ((cur == '<') && (next == '!') &&
5648 (avail < 9)) {
5649 goto done;
5650 } else {
5651 ctxt->instate = XML_PARSER_CONTENT;
5652#ifdef DEBUG_PUSH
5653 xmlGenericError(xmlGenericErrorContext,
5654 "HPP: entering START_TAG\n");
5655#endif
5656 }
5657 break;
5658 case XML_PARSER_PROLOG:
5659 SKIP_BLANKS;
5660 if (in->buf == NULL)
5661 avail = in->length - (in->cur - in->base);
5662 else
5663 avail = (ptrdiff_t)xmlBufUse(in->buf->buffer) -
5664 (in->cur - in->base);
5665 if (avail < 2)
5666 goto done;
5667 cur = in->cur[0];
5668 next = in->cur[1];
5669 if ((cur == '<') && (next == '!') &&
5670 (in->cur[2] == '-') && (in->cur[3] == '-')) {
5671 if ((!terminate) && (htmlParseLookupCommentEnd(ctxt) < 0))
5672 goto done;
5673#ifdef DEBUG_PUSH
5674 xmlGenericError(xmlGenericErrorContext,
5675 "HPP: Parsing Comment\n");
5676#endif
5677 htmlParseComment(ctxt);
5678 ctxt->instate = XML_PARSER_PROLOG;
5679 } else if ((cur == '<') && (next == '?')) {
5680 if ((!terminate) &&
5681 (htmlParseLookupSequence(ctxt, '>', 0, 0, 0) < 0))
5682 goto done;
5683#ifdef DEBUG_PUSH
5684 xmlGenericError(xmlGenericErrorContext,
5685 "HPP: Parsing PI\n");
5686#endif
5687 htmlParsePI(ctxt);
5688 ctxt->instate = XML_PARSER_PROLOG;
5689 } else if ((cur == '<') && (next == '!') &&
5690 (avail < 4)) {
5691 goto done;
5692 } else {
5693 ctxt->instate = XML_PARSER_CONTENT;
5694#ifdef DEBUG_PUSH
5695 xmlGenericError(xmlGenericErrorContext,
5696 "HPP: entering START_TAG\n");
5697#endif
5698 }
5699 break;
5700 case XML_PARSER_EPILOG:
5701 if (in->buf == NULL)
5702 avail = in->length - (in->cur - in->base);
5703 else
5704 avail = (ptrdiff_t)xmlBufUse(in->buf->buffer) -
5705 (in->cur - in->base);
5706 if (avail < 1)
5707 goto done;
5708 cur = in->cur[0];
5709 if (IS_BLANK_CH(cur)) {
5710 htmlParseCharData(ctxt);
5711 goto done;
5712 }
5713 if (avail < 2)
5714 goto done;
5715 next = in->cur[1];
5716 if ((cur == '<') && (next == '!') &&
5717 (in->cur[2] == '-') && (in->cur[3] == '-')) {
5718 if ((!terminate) && (htmlParseLookupCommentEnd(ctxt) < 0))
5719 goto done;
5720#ifdef DEBUG_PUSH
5721 xmlGenericError(xmlGenericErrorContext,
5722 "HPP: Parsing Comment\n");
5723#endif
5724 htmlParseComment(ctxt);
5725 ctxt->instate = XML_PARSER_EPILOG;
5726 } else if ((cur == '<') && (next == '?')) {
5727 if ((!terminate) &&
5728 (htmlParseLookupSequence(ctxt, '>', 0, 0, 0) < 0))
5729 goto done;
5730#ifdef DEBUG_PUSH
5731 xmlGenericError(xmlGenericErrorContext,
5732 "HPP: Parsing PI\n");
5733#endif
5734 htmlParsePI(ctxt);
5735 ctxt->instate = XML_PARSER_EPILOG;
5736 } else if ((cur == '<') && (next == '!') &&
5737 (avail < 4)) {
5738 goto done;
5739 } else {
5740 ctxt->errNo = XML_ERR_DOCUMENT_END;
5741 ctxt->wellFormed = 0;
5742 ctxt->instate = XML_PARSER_EOF;
5743#ifdef DEBUG_PUSH
5744 xmlGenericError(xmlGenericErrorContext,
5745 "HPP: entering EOF\n");
5746#endif
5747 if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
5748 ctxt->sax->endDocument(ctxt->userData);
5749 goto done;
5750 }
5751 break;
5752 case XML_PARSER_START_TAG: {
5753 const xmlChar *name;
5754 int failed;
5755 const htmlElemDesc * info;
5756
5757 /*
5758 * no chars in buffer
5759 */
5760 if (avail < 1)
5761 goto done;
5762 /*
5763 * not enough chars in buffer
5764 */
5765 if (avail < 2) {
5766 if (!terminate)
5767 goto done;
5768 else
5769 next = ' ';
5770 } else {
5771 next = in->cur[1];
5772 }
5773 cur = in->cur[0];
5774 if (cur != '<') {
5775 ctxt->instate = XML_PARSER_CONTENT;
5776#ifdef DEBUG_PUSH
5777 xmlGenericError(xmlGenericErrorContext,
5778 "HPP: entering CONTENT\n");
5779#endif
5780 break;
5781 }
5782 if (next == '/') {
5783 ctxt->instate = XML_PARSER_END_TAG;
5784 ctxt->checkIndex = 0;
5785#ifdef DEBUG_PUSH
5786 xmlGenericError(xmlGenericErrorContext,
5787 "HPP: entering END_TAG\n");
5788#endif
5789 break;
5790 }
5791 if ((!terminate) &&
5792 (htmlParseLookupSequence(ctxt, '>', 0, 0, 1) < 0))
5793 goto done;
5794
5795 /* Capture start position */
5796 if (ctxt->record_info) {
5797 node_info.begin_pos = ctxt->input->consumed +
5798 (CUR_PTR - ctxt->input->base);
5799 node_info.begin_line = ctxt->input->line;
5800 }
5801
5802
5803 failed = htmlParseStartTag(ctxt);
5804 name = ctxt->name;
5805 if ((failed == -1) ||
5806 (name == NULL)) {
5807 if (CUR == '>')
5808 NEXT;
5809 break;
5810 }
5811
5812 /*
5813 * Lookup the info for that element.
5814 */
5815 info = htmlTagLookup(name);
5816 if (info == NULL) {
5817 htmlParseErr(ctxt, XML_HTML_UNKNOWN_TAG,
5818 "Tag %s invalid\n", name, NULL);
5819 }
5820
5821 /*
5822 * Check for an Empty Element labeled the XML/SGML way
5823 */
5824 if ((CUR == '/') && (NXT(1) == '>')) {
5825 SKIP(2);
5826 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
5827 ctxt->sax->endElement(ctxt->userData, name);
5828 htmlnamePop(ctxt);
5829 ctxt->instate = XML_PARSER_CONTENT;
5830#ifdef DEBUG_PUSH
5831 xmlGenericError(xmlGenericErrorContext,
5832 "HPP: entering CONTENT\n");
5833#endif
5834 break;
5835 }
5836
5837 if (CUR == '>') {
5838 NEXT;
5839 } else {
5840 htmlParseErr(ctxt, XML_ERR_GT_REQUIRED,
5841 "Couldn't find end of Start Tag %s\n",
5842 name, NULL);
5843
5844 /*
5845 * end of parsing of this node.
5846 */
5847 if (xmlStrEqual(name, ctxt->name)) {
5848 nodePop(ctxt);
5849 htmlnamePop(ctxt);
5850 }
5851
5852 if (ctxt->record_info)
5853 htmlNodeInfoPush(ctxt, &node_info);
5854
5855 ctxt->instate = XML_PARSER_CONTENT;
5856#ifdef DEBUG_PUSH
5857 xmlGenericError(xmlGenericErrorContext,
5858 "HPP: entering CONTENT\n");
5859#endif
5860 break;
5861 }
5862
5863 /*
5864 * Check for an Empty Element from DTD definition
5865 */
5866 if ((info != NULL) && (info->empty)) {
5867 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
5868 ctxt->sax->endElement(ctxt->userData, name);
5869 htmlnamePop(ctxt);
5870 }
5871
5872 if (ctxt->record_info)
5873 htmlNodeInfoPush(ctxt, &node_info);
5874
5875 ctxt->instate = XML_PARSER_CONTENT;
5876#ifdef DEBUG_PUSH
5877 xmlGenericError(xmlGenericErrorContext,
5878 "HPP: entering CONTENT\n");
5879#endif
5880 break;
5881 }
5882 case XML_PARSER_CONTENT: {
5883 xmlChar chr[2] = { 0, 0 };
5884
5885 /*
5886 * Handle preparsed entities and charRef
5887 */
5888 if (ctxt->token != 0) {
5889 chr[0] = (xmlChar) ctxt->token;
5890 htmlCheckParagraph(ctxt);
5891 if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
5892 ctxt->sax->characters(ctxt->userData, chr, 1);
5893 ctxt->token = 0;
5894 ctxt->checkIndex = 0;
5895 }
5896 if ((avail == 1) && (terminate)) {
5897 cur = in->cur[0];
5898 if ((cur != '<') && (cur != '&')) {
5899 if (ctxt->sax != NULL) {
5900 chr[0] = cur;
5901 if (IS_BLANK_CH(cur)) {
5902 if (ctxt->keepBlanks) {
5903 if (ctxt->sax->characters != NULL)
5904 ctxt->sax->characters(
5905 ctxt->userData, chr, 1);
5906 } else {
5907 if (ctxt->sax->ignorableWhitespace != NULL)
5908 ctxt->sax->ignorableWhitespace(
5909 ctxt->userData, chr, 1);
5910 }
5911 } else {
5912 htmlCheckParagraph(ctxt);
5913 if (ctxt->sax->characters != NULL)
5914 ctxt->sax->characters(
5915 ctxt->userData, chr, 1);
5916 }
5917 }
5918 ctxt->token = 0;
5919 ctxt->checkIndex = 0;
5920 in->cur++;
5921 break;
5922 }
5923 }
5924 if (avail < 2)
5925 goto done;
5926 cur = in->cur[0];
5927 next = in->cur[1];
5928 if ((xmlStrEqual(ctxt->name, BAD_CAST"script")) ||
5929 (xmlStrEqual(ctxt->name, BAD_CAST"style"))) {
5930 /*
5931 * Handle SCRIPT/STYLE separately
5932 */
5933 if (!terminate) {
5934 int idx;
5935 xmlChar val;
5936
5937 idx = htmlParseLookupSequence(ctxt, '<', '/', 0, 0);
5938 if (idx < 0)
5939 goto done;
5940 val = in->cur[idx + 2];
5941 if (val == 0) /* bad cut of input */
5942 goto done;
5943 }
5944 htmlParseScript(ctxt);
5945 if ((cur == '<') && (next == '/')) {
5946 ctxt->instate = XML_PARSER_END_TAG;
5947 ctxt->checkIndex = 0;
5948#ifdef DEBUG_PUSH
5949 xmlGenericError(xmlGenericErrorContext,
5950 "HPP: entering END_TAG\n");
5951#endif
5952 break;
5953 }
5954 } else {
5955 /*
5956 * Sometimes DOCTYPE arrives in the middle of the document
5957 */
5958 if ((cur == '<') && (next == '!') &&
5959 (UPP(2) == 'D') && (UPP(3) == 'O') &&
5960 (UPP(4) == 'C') && (UPP(5) == 'T') &&
5961 (UPP(6) == 'Y') && (UPP(7) == 'P') &&
5962 (UPP(8) == 'E')) {
5963 if ((!terminate) &&
5964 (htmlParseLookupSequence(ctxt, '>', 0, 0, 1) < 0))
5965 goto done;
5966 htmlParseErr(ctxt, XML_HTML_STRUCURE_ERROR,
5967 "Misplaced DOCTYPE declaration\n",
5968 BAD_CAST "DOCTYPE" , NULL);
5969 htmlParseDocTypeDecl(ctxt);
5970 } else if ((cur == '<') && (next == '!') &&
5971 (in->cur[2] == '-') && (in->cur[3] == '-')) {
5972 if ((!terminate) && (htmlParseLookupCommentEnd(ctxt) < 0))
5973 goto done;
5974#ifdef DEBUG_PUSH
5975 xmlGenericError(xmlGenericErrorContext,
5976 "HPP: Parsing Comment\n");
5977#endif
5978 htmlParseComment(ctxt);
5979 ctxt->instate = XML_PARSER_CONTENT;
5980 } else if ((cur == '<') && (next == '?')) {
5981 if ((!terminate) &&
5982 (htmlParseLookupSequence(ctxt, '>', 0, 0, 0) < 0))
5983 goto done;
5984#ifdef DEBUG_PUSH
5985 xmlGenericError(xmlGenericErrorContext,
5986 "HPP: Parsing PI\n");
5987#endif
5988 htmlParsePI(ctxt);
5989 ctxt->instate = XML_PARSER_CONTENT;
5990 } else if ((cur == '<') && (next == '!') && (avail < 4)) {
5991 goto done;
5992 } else if ((cur == '<') && (next == '/')) {
5993 ctxt->instate = XML_PARSER_END_TAG;
5994 ctxt->checkIndex = 0;
5995#ifdef DEBUG_PUSH
5996 xmlGenericError(xmlGenericErrorContext,
5997 "HPP: entering END_TAG\n");
5998#endif
5999 break;
6000 } else if ((cur == '<') && IS_ASCII_LETTER(next)) {
6001 if ((!terminate) && (next == 0))
6002 goto done;
6003 ctxt->instate = XML_PARSER_START_TAG;
6004 ctxt->checkIndex = 0;
6005#ifdef DEBUG_PUSH
6006 xmlGenericError(xmlGenericErrorContext,
6007 "HPP: entering START_TAG\n");
6008#endif
6009 break;
6010 } else if (cur == '<') {
6011 if ((ctxt->sax != NULL) && (!ctxt->disableSAX) &&
6012 (ctxt->sax->characters != NULL))
6013 ctxt->sax->characters(ctxt->userData,
6014 BAD_CAST "<", 1);
6015 NEXT;
6016 } else {
6017 /*
6018 * check that the text sequence is complete
6019 * before handing out the data to the parser
6020 * to avoid problems with erroneous end of
6021 * data detection.
6022 */
6023 if ((!terminate) &&
6024 (htmlParseLookupSequence(ctxt, '<', 0, 0, 0) < 0))
6025 goto done;
6026 ctxt->checkIndex = 0;
6027#ifdef DEBUG_PUSH
6028 xmlGenericError(xmlGenericErrorContext,
6029 "HPP: Parsing char data\n");
6030#endif
6031 while ((ctxt->instate != XML_PARSER_EOF) &&
6032 (cur != '<') && (in->cur < in->end)) {
6033 if (cur == '&') {
6034 htmlParseReference(ctxt);
6035 } else {
6036 htmlParseCharData(ctxt);
6037 }
6038 cur = in->cur[0];
6039 }
6040 }
6041 }
6042
6043 break;
6044 }
6045 case XML_PARSER_END_TAG:
6046 if (avail < 2)
6047 goto done;
6048 if ((!terminate) &&
6049 (htmlParseLookupSequence(ctxt, '>', 0, 0, 0) < 0))
6050 goto done;
6051 htmlParseEndTag(ctxt);
6052 if (ctxt->nameNr == 0) {
6053 ctxt->instate = XML_PARSER_EPILOG;
6054 } else {
6055 ctxt->instate = XML_PARSER_CONTENT;
6056 }
6057 ctxt->checkIndex = 0;
6058#ifdef DEBUG_PUSH
6059 xmlGenericError(xmlGenericErrorContext,
6060 "HPP: entering CONTENT\n");
6061#endif
6062 break;
6063 case XML_PARSER_CDATA_SECTION:
6064 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
6065 "HPP: internal error, state == CDATA\n",
6066 NULL, NULL);
6067 ctxt->instate = XML_PARSER_CONTENT;
6068 ctxt->checkIndex = 0;
6069#ifdef DEBUG_PUSH
6070 xmlGenericError(xmlGenericErrorContext,
6071 "HPP: entering CONTENT\n");
6072#endif
6073 break;
6074 case XML_PARSER_DTD:
6075 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
6076 "HPP: internal error, state == DTD\n",
6077 NULL, NULL);
6078 ctxt->instate = XML_PARSER_CONTENT;
6079 ctxt->checkIndex = 0;
6080#ifdef DEBUG_PUSH
6081 xmlGenericError(xmlGenericErrorContext,
6082 "HPP: entering CONTENT\n");
6083#endif
6084 break;
6085 case XML_PARSER_COMMENT:
6086 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
6087 "HPP: internal error, state == COMMENT\n",
6088 NULL, NULL);
6089 ctxt->instate = XML_PARSER_CONTENT;
6090 ctxt->checkIndex = 0;
6091#ifdef DEBUG_PUSH
6092 xmlGenericError(xmlGenericErrorContext,
6093 "HPP: entering CONTENT\n");
6094#endif
6095 break;
6096 case XML_PARSER_PI:
6097 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
6098 "HPP: internal error, state == PI\n",
6099 NULL, NULL);
6100 ctxt->instate = XML_PARSER_CONTENT;
6101 ctxt->checkIndex = 0;
6102#ifdef DEBUG_PUSH
6103 xmlGenericError(xmlGenericErrorContext,
6104 "HPP: entering CONTENT\n");
6105#endif
6106 break;
6107 case XML_PARSER_ENTITY_DECL:
6108 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
6109 "HPP: internal error, state == ENTITY_DECL\n",
6110 NULL, NULL);
6111 ctxt->instate = XML_PARSER_CONTENT;
6112 ctxt->checkIndex = 0;
6113#ifdef DEBUG_PUSH
6114 xmlGenericError(xmlGenericErrorContext,
6115 "HPP: entering CONTENT\n");
6116#endif
6117 break;
6118 case XML_PARSER_ENTITY_VALUE:
6119 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
6120 "HPP: internal error, state == ENTITY_VALUE\n",
6121 NULL, NULL);
6122 ctxt->instate = XML_PARSER_CONTENT;
6123 ctxt->checkIndex = 0;
6124#ifdef DEBUG_PUSH
6125 xmlGenericError(xmlGenericErrorContext,
6126 "HPP: entering DTD\n");
6127#endif
6128 break;
6129 case XML_PARSER_ATTRIBUTE_VALUE:
6130 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
6131 "HPP: internal error, state == ATTRIBUTE_VALUE\n",
6132 NULL, NULL);
6133 ctxt->instate = XML_PARSER_START_TAG;
6134 ctxt->checkIndex = 0;
6135#ifdef DEBUG_PUSH
6136 xmlGenericError(xmlGenericErrorContext,
6137 "HPP: entering START_TAG\n");
6138#endif
6139 break;
6140 case XML_PARSER_SYSTEM_LITERAL:
6141 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
6142 "HPP: internal error, state == XML_PARSER_SYSTEM_LITERAL\n",
6143 NULL, NULL);
6144 ctxt->instate = XML_PARSER_CONTENT;
6145 ctxt->checkIndex = 0;
6146#ifdef DEBUG_PUSH
6147 xmlGenericError(xmlGenericErrorContext,
6148 "HPP: entering CONTENT\n");
6149#endif
6150 break;
6151 case XML_PARSER_IGNORE:
6152 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
6153 "HPP: internal error, state == XML_PARSER_IGNORE\n",
6154 NULL, NULL);
6155 ctxt->instate = XML_PARSER_CONTENT;
6156 ctxt->checkIndex = 0;
6157#ifdef DEBUG_PUSH
6158 xmlGenericError(xmlGenericErrorContext,
6159 "HPP: entering CONTENT\n");
6160#endif
6161 break;
6162 case XML_PARSER_PUBLIC_LITERAL:
6163 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
6164 "HPP: internal error, state == XML_PARSER_LITERAL\n",
6165 NULL, NULL);
6166 ctxt->instate = XML_PARSER_CONTENT;
6167 ctxt->checkIndex = 0;
6168#ifdef DEBUG_PUSH
6169 xmlGenericError(xmlGenericErrorContext,
6170 "HPP: entering CONTENT\n");
6171#endif
6172 break;
6173
6174 }
6175 }
6176done:
6177 if ((avail == 0) && (terminate)) {
6178 htmlAutoCloseOnEnd(ctxt);
6179 if ((ctxt->nameNr == 0) && (ctxt->instate != XML_PARSER_EOF)) {
6180 /*
6181 * SAX: end of the document processing.
6182 */
6183 ctxt->instate = XML_PARSER_EOF;
6184 if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
6185 ctxt->sax->endDocument(ctxt->userData);
6186 }
6187 }
6188 if ((!(ctxt->options & HTML_PARSE_NODEFDTD)) && (ctxt->myDoc != NULL) &&
6189 ((terminate) || (ctxt->instate == XML_PARSER_EOF) ||
6190 (ctxt->instate == XML_PARSER_EPILOG))) {
6191 xmlDtdPtr dtd;
6192 dtd = xmlGetIntSubset(ctxt->myDoc);
6193 if (dtd == NULL)
6194 ctxt->myDoc->intSubset =
6195 xmlCreateIntSubset(ctxt->myDoc, BAD_CAST "html",
6196 BAD_CAST "-//W3C//DTD HTML 4.0 Transitional//EN",
6197 BAD_CAST "http://www.w3.org/TR/REC-html40/loose.dtd");
6198 }
6199#ifdef DEBUG_PUSH
6200 xmlGenericError(xmlGenericErrorContext, "HPP: done %d\n", ret);
6201#endif
6202 return(ret);
6203}
6204
6205/**
6206 * htmlParseChunk:
6207 * @ctxt: an HTML parser context
6208 * @chunk: an char array
6209 * @size: the size in byte of the chunk
6210 * @terminate: last chunk indicator
6211 *
6212 * Parse a Chunk of memory
6213 *
6214 * Returns zero if no error, the xmlParserErrors otherwise.
6215 */
6216int
6217htmlParseChunk(htmlParserCtxtPtr ctxt, const char *chunk, int size,
6218 int terminate) {
6219 if ((ctxt == NULL) || (ctxt->input == NULL)) {
6220 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
6221 "htmlParseChunk: context error\n", NULL, NULL);
6222 return(XML_ERR_INTERNAL_ERROR);
6223 }
6224 if ((size > 0) && (chunk != NULL) && (ctxt->input != NULL) &&
6225 (ctxt->input->buf != NULL) && (ctxt->instate != XML_PARSER_EOF)) {
6226 size_t base = xmlBufGetInputBase(ctxt->input->buf->buffer, ctxt->input);
6227 size_t cur = ctxt->input->cur - ctxt->input->base;
6228 int res;
6229
6230 res = xmlParserInputBufferPush(ctxt->input->buf, size, chunk);
6231 xmlBufSetInputBaseCur(ctxt->input->buf->buffer, ctxt->input, base, cur);
6232 if (res < 0) {
6233 ctxt->errNo = XML_PARSER_EOF;
6234 ctxt->disableSAX = 1;
6235 return (XML_PARSER_EOF);
6236 }
6237#ifdef DEBUG_PUSH
6238 xmlGenericError(xmlGenericErrorContext, "HPP: pushed %d\n", size);
6239#endif
6240
6241#if 0
6242 if ((terminate) || (ctxt->input->buf->buffer->use > 80))
6243 htmlParseTryOrFinish(ctxt, terminate);
6244#endif
6245 } else if (ctxt->instate != XML_PARSER_EOF) {
6246 if ((ctxt->input != NULL) && ctxt->input->buf != NULL) {
6247 xmlParserInputBufferPtr in = ctxt->input->buf;
6248 if ((in->encoder != NULL) && (in->buffer != NULL) &&
6249 (in->raw != NULL)) {
6250 int nbchars;
6251 size_t base = xmlBufGetInputBase(in->buffer, ctxt->input);
6252 size_t current = ctxt->input->cur - ctxt->input->base;
6253
6254 nbchars = xmlCharEncInput(in, terminate);
6255 xmlBufSetInputBaseCur(in->buffer, ctxt->input, base, current);
6256 if (nbchars < 0) {
6257 htmlParseErr(ctxt, XML_ERR_INVALID_ENCODING,
6258 "encoder error\n", NULL, NULL);
6259 return(XML_ERR_INVALID_ENCODING);
6260 }
6261 }
6262 }
6263 }
6264 htmlParseTryOrFinish(ctxt, terminate);
6265 if (terminate) {
6266 if ((ctxt->instate != XML_PARSER_EOF) &&
6267 (ctxt->instate != XML_PARSER_EPILOG) &&
6268 (ctxt->instate != XML_PARSER_MISC)) {
6269 ctxt->errNo = XML_ERR_DOCUMENT_END;
6270 ctxt->wellFormed = 0;
6271 }
6272 if (ctxt->instate != XML_PARSER_EOF) {
6273 if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
6274 ctxt->sax->endDocument(ctxt->userData);
6275 }
6276 ctxt->instate = XML_PARSER_EOF;
6277 }
6278 return((xmlParserErrors) ctxt->errNo);
6279}
6280
6281/************************************************************************
6282 * *
6283 * User entry points *
6284 * *
6285 ************************************************************************/
6286
6287/**
6288 * htmlCreatePushParserCtxt:
6289 * @sax: a SAX handler
6290 * @user_data: The user data returned on SAX callbacks
6291 * @chunk: a pointer to an array of chars
6292 * @size: number of chars in the array
6293 * @filename: an optional file name or URI
6294 * @enc: an optional encoding
6295 *
6296 * Create a parser context for using the HTML parser in push mode
6297 * The value of @filename is used for fetching external entities
6298 * and error/warning reports.
6299 *
6300 * Returns the new parser context or NULL
6301 */
6302htmlParserCtxtPtr
6303htmlCreatePushParserCtxt(htmlSAXHandlerPtr sax, void *user_data,
6304 const char *chunk, int size, const char *filename,
6305 xmlCharEncoding enc) {
6306 htmlParserCtxtPtr ctxt;
6307 htmlParserInputPtr inputStream;
6308 xmlParserInputBufferPtr buf;
6309
6310 xmlInitParser();
6311
6312 buf = xmlAllocParserInputBuffer(enc);
6313 if (buf == NULL) return(NULL);
6314
6315 ctxt = htmlNewParserCtxt();
6316 if (ctxt == NULL) {
6317 xmlFreeParserInputBuffer(buf);
6318 return(NULL);
6319 }
6320 if(enc==XML_CHAR_ENCODING_UTF8 || buf->encoder)
6321 ctxt->charset=XML_CHAR_ENCODING_UTF8;
6322 if (sax != NULL) {
6323 if (ctxt->sax != (xmlSAXHandlerPtr) &htmlDefaultSAXHandler)
6324 xmlFree(ctxt->sax);
6325 ctxt->sax = (htmlSAXHandlerPtr) xmlMalloc(sizeof(htmlSAXHandler));
6326 if (ctxt->sax == NULL) {
6327 xmlFree(buf);
6328 xmlFree(ctxt);
6329 return(NULL);
6330 }
6331 memcpy(ctxt->sax, sax, sizeof(htmlSAXHandler));
6332 if (user_data != NULL)
6333 ctxt->userData = user_data;
6334 }
6335 if (filename == NULL) {
6336 ctxt->directory = NULL;
6337 } else {
6338 ctxt->directory = xmlParserGetDirectory(filename);
6339 }
6340
6341 inputStream = htmlNewInputStream(ctxt);
6342 if (inputStream == NULL) {
6343 xmlFreeParserCtxt(ctxt);
6344 xmlFree(buf);
6345 return(NULL);
6346 }
6347
6348 if (filename == NULL)
6349 inputStream->filename = NULL;
6350 else
6351 inputStream->filename = (char *)
6352 xmlCanonicPath((const xmlChar *) filename);
6353 inputStream->buf = buf;
6354 xmlBufResetInput(buf->buffer, inputStream);
6355
6356 inputPush(ctxt, inputStream);
6357
6358 if ((size > 0) && (chunk != NULL) && (ctxt->input != NULL) &&
6359 (ctxt->input->buf != NULL)) {
6360 size_t base = xmlBufGetInputBase(ctxt->input->buf->buffer, ctxt->input);
6361 size_t cur = ctxt->input->cur - ctxt->input->base;
6362
6363 xmlParserInputBufferPush(ctxt->input->buf, size, chunk);
6364
6365 xmlBufSetInputBaseCur(ctxt->input->buf->buffer, ctxt->input, base, cur);
6366#ifdef DEBUG_PUSH
6367 xmlGenericError(xmlGenericErrorContext, "HPP: pushed %d\n", size);
6368#endif
6369 }
6370 ctxt->progressive = 1;
6371
6372 return(ctxt);
6373}
6374#endif /* LIBXML_PUSH_ENABLED */
6375
6376/**
6377 * htmlSAXParseDoc:
6378 * @cur: a pointer to an array of xmlChar
6379 * @encoding: a free form C string describing the HTML document encoding, or NULL
6380 * @sax: the SAX handler block
6381 * @userData: if using SAX, this pointer will be provided on callbacks.
6382 *
6383 * Parse an HTML in-memory document. If sax is not NULL, use the SAX callbacks
6384 * to handle parse events. If sax is NULL, fallback to the default DOM
6385 * behavior and return a tree.
6386 *
6387 * Returns the resulting document tree unless SAX is NULL or the document is
6388 * not well formed.
6389 */
6390
6391htmlDocPtr
6392htmlSAXParseDoc(const xmlChar *cur, const char *encoding,
6393 htmlSAXHandlerPtr sax, void *userData) {
6394 htmlDocPtr ret;
6395 htmlParserCtxtPtr ctxt;
6396
6397 xmlInitParser();
6398
6399 if (cur == NULL) return(NULL);
6400
6401
6402 ctxt = htmlCreateDocParserCtxt(cur, encoding);
6403 if (ctxt == NULL) return(NULL);
6404 if (sax != NULL) {
6405 if (ctxt->sax != NULL) xmlFree (ctxt->sax);
6406 ctxt->sax = sax;
6407 ctxt->userData = userData;
6408 }
6409
6410 htmlParseDocument(ctxt);
6411 ret = ctxt->myDoc;
6412 if (sax != NULL) {
6413 ctxt->sax = NULL;
6414 ctxt->userData = NULL;
6415 }
6416 htmlFreeParserCtxt(ctxt);
6417
6418 return(ret);
6419}
6420
6421/**
6422 * htmlParseDoc:
6423 * @cur: a pointer to an array of xmlChar
6424 * @encoding: a free form C string describing the HTML document encoding, or NULL
6425 *
6426 * parse an HTML in-memory document and build a tree.
6427 *
6428 * Returns the resulting document tree
6429 */
6430
6431htmlDocPtr
6432htmlParseDoc(const xmlChar *cur, const char *encoding) {
6433 return(htmlSAXParseDoc(cur, encoding, NULL, NULL));
6434}
6435
6436
6437/**
6438 * htmlCreateFileParserCtxt:
6439 * @filename: the filename
6440 * @encoding: a free form C string describing the HTML document encoding, or NULL
6441 *
6442 * Create a parser context for a file content.
6443 * Automatic support for ZLIB/Compress compressed document is provided
6444 * by default if found at compile-time.
6445 *
6446 * Returns the new parser context or NULL
6447 */
6448htmlParserCtxtPtr
6449htmlCreateFileParserCtxt(const char *filename, const char *encoding)
6450{
6451 htmlParserCtxtPtr ctxt;
6452 htmlParserInputPtr inputStream;
6453 char *canonicFilename;
6454 /* htmlCharEncoding enc; */
6455 xmlChar *content, *content_line = (xmlChar *) "charset=";
6456
6457 if (filename == NULL)
6458 return(NULL);
6459
6460 ctxt = htmlNewParserCtxt();
6461 if (ctxt == NULL) {
6462 return(NULL);
6463 }
6464 canonicFilename = (char *) xmlCanonicPath((const xmlChar *) filename);
6465 if (canonicFilename == NULL) {
6466#ifdef LIBXML_SAX1_ENABLED
6467 if (xmlDefaultSAXHandler.error != NULL) {
6468 xmlDefaultSAXHandler.error(NULL, "out of memory\n");
6469 }
6470#endif
6471 xmlFreeParserCtxt(ctxt);
6472 return(NULL);
6473 }
6474
6475 inputStream = xmlLoadExternalEntity(canonicFilename, NULL, ctxt);
6476 xmlFree(canonicFilename);
6477 if (inputStream == NULL) {
6478 xmlFreeParserCtxt(ctxt);
6479 return(NULL);
6480 }
6481
6482 inputPush(ctxt, inputStream);
6483
6484 /* set encoding */
6485 if (encoding) {
6486 size_t l = strlen(encoding);
6487
6488 if (l < 1000) {
6489 content = xmlMallocAtomic (xmlStrlen(content_line) + l + 1);
6490 if (content) {
6491 strcpy ((char *)content, (char *)content_line);
6492 strcat ((char *)content, (char *)encoding);
6493 htmlCheckEncoding (ctxt, content);
6494 xmlFree (content);
6495 }
6496 }
6497 }
6498
6499 return(ctxt);
6500}
6501
6502/**
6503 * htmlSAXParseFile:
6504 * @filename: the filename
6505 * @encoding: a free form C string describing the HTML document encoding, or NULL
6506 * @sax: the SAX handler block
6507 * @userData: if using SAX, this pointer will be provided on callbacks.
6508 *
6509 * parse an HTML file and build a tree. Automatic support for ZLIB/Compress
6510 * compressed document is provided by default if found at compile-time.
6511 * It use the given SAX function block to handle the parsing callback.
6512 * If sax is NULL, fallback to the default DOM tree building routines.
6513 *
6514 * Returns the resulting document tree unless SAX is NULL or the document is
6515 * not well formed.
6516 */
6517
6518htmlDocPtr
6519htmlSAXParseFile(const char *filename, const char *encoding, htmlSAXHandlerPtr sax,
6520 void *userData) {
6521 htmlDocPtr ret;
6522 htmlParserCtxtPtr ctxt;
6523 htmlSAXHandlerPtr oldsax = NULL;
6524
6525 xmlInitParser();
6526
6527 ctxt = htmlCreateFileParserCtxt(filename, encoding);
6528 if (ctxt == NULL) return(NULL);
6529 if (sax != NULL) {
6530 oldsax = ctxt->sax;
6531 ctxt->sax = sax;
6532 ctxt->userData = userData;
6533 }
6534
6535 htmlParseDocument(ctxt);
6536
6537 ret = ctxt->myDoc;
6538 if (sax != NULL) {
6539 ctxt->sax = oldsax;
6540 ctxt->userData = NULL;
6541 }
6542 htmlFreeParserCtxt(ctxt);
6543
6544 return(ret);
6545}
6546
6547/**
6548 * htmlParseFile:
6549 * @filename: the filename
6550 * @encoding: a free form C string describing the HTML document encoding, or NULL
6551 *
6552 * parse an HTML file and build a tree. Automatic support for ZLIB/Compress
6553 * compressed document is provided by default if found at compile-time.
6554 *
6555 * Returns the resulting document tree
6556 */
6557
6558htmlDocPtr
6559htmlParseFile(const char *filename, const char *encoding) {
6560 return(htmlSAXParseFile(filename, encoding, NULL, NULL));
6561}
6562
6563/**
6564 * htmlHandleOmittedElem:
6565 * @val: int 0 or 1
6566 *
6567 * Set and return the previous value for handling HTML omitted tags.
6568 *
6569 * Returns the last value for 0 for no handling, 1 for auto insertion.
6570 */
6571
6572int
6573htmlHandleOmittedElem(int val) {
6574 int old = htmlOmittedDefaultValue;
6575
6576 htmlOmittedDefaultValue = val;
6577 return(old);
6578}
6579
6580/**
6581 * htmlElementAllowedHere:
6582 * @parent: HTML parent element
6583 * @elt: HTML element
6584 *
6585 * Checks whether an HTML element may be a direct child of a parent element.
6586 * Note - doesn't check for deprecated elements
6587 *
6588 * Returns 1 if allowed; 0 otherwise.
6589 */
6590int
6591htmlElementAllowedHere(const htmlElemDesc* parent, const xmlChar* elt) {
6592 const char** p ;
6593
6594 if ( ! elt || ! parent || ! parent->subelts )
6595 return 0 ;
6596
6597 for ( p = parent->subelts; *p; ++p )
6598 if ( !xmlStrcmp((const xmlChar *)*p, elt) )
6599 return 1 ;
6600
6601 return 0 ;
6602}
6603/**
6604 * htmlElementStatusHere:
6605 * @parent: HTML parent element
6606 * @elt: HTML element
6607 *
6608 * Checks whether an HTML element may be a direct child of a parent element.
6609 * and if so whether it is valid or deprecated.
6610 *
6611 * Returns one of HTML_VALID, HTML_DEPRECATED, HTML_INVALID
6612 */
6613htmlStatus
6614htmlElementStatusHere(const htmlElemDesc* parent, const htmlElemDesc* elt) {
6615 if ( ! parent || ! elt )
6616 return HTML_INVALID ;
6617 if ( ! htmlElementAllowedHere(parent, (const xmlChar*) elt->name ) )
6618 return HTML_INVALID ;
6619
6620 return ( elt->dtd == 0 ) ? HTML_VALID : HTML_DEPRECATED ;
6621}
6622/**
6623 * htmlAttrAllowed:
6624 * @elt: HTML element
6625 * @attr: HTML attribute
6626 * @legacy: whether to allow deprecated attributes
6627 *
6628 * Checks whether an attribute is valid for an element
6629 * Has full knowledge of Required and Deprecated attributes
6630 *
6631 * Returns one of HTML_REQUIRED, HTML_VALID, HTML_DEPRECATED, HTML_INVALID
6632 */
6633htmlStatus
6634htmlAttrAllowed(const htmlElemDesc* elt, const xmlChar* attr, int legacy) {
6635 const char** p ;
6636
6637 if ( !elt || ! attr )
6638 return HTML_INVALID ;
6639
6640 if ( elt->attrs_req )
6641 for ( p = elt->attrs_req; *p; ++p)
6642 if ( !xmlStrcmp((const xmlChar*)*p, attr) )
6643 return HTML_REQUIRED ;
6644
6645 if ( elt->attrs_opt )
6646 for ( p = elt->attrs_opt; *p; ++p)
6647 if ( !xmlStrcmp((const xmlChar*)*p, attr) )
6648 return HTML_VALID ;
6649
6650 if ( legacy && elt->attrs_depr )
6651 for ( p = elt->attrs_depr; *p; ++p)
6652 if ( !xmlStrcmp((const xmlChar*)*p, attr) )
6653 return HTML_DEPRECATED ;
6654
6655 return HTML_INVALID ;
6656}
6657/**
6658 * htmlNodeStatus:
6659 * @node: an htmlNodePtr in a tree
6660 * @legacy: whether to allow deprecated elements (YES is faster here
6661 * for Element nodes)
6662 *
6663 * Checks whether the tree node is valid. Experimental (the author
6664 * only uses the HTML enhancements in a SAX parser)
6665 *
6666 * Return: for Element nodes, a return from htmlElementAllowedHere (if
6667 * legacy allowed) or htmlElementStatusHere (otherwise).
6668 * for Attribute nodes, a return from htmlAttrAllowed
6669 * for other nodes, HTML_NA (no checks performed)
6670 */
6671htmlStatus
6672htmlNodeStatus(const htmlNodePtr node, int legacy) {
6673 if ( ! node )
6674 return HTML_INVALID ;
6675
6676 switch ( node->type ) {
6677 case XML_ELEMENT_NODE:
6678 return legacy
6679 ? ( htmlElementAllowedHere (
6680 htmlTagLookup(node->parent->name) , node->name
6681 ) ? HTML_VALID : HTML_INVALID )
6682 : htmlElementStatusHere(
6683 htmlTagLookup(node->parent->name) ,
6684 htmlTagLookup(node->name) )
6685 ;
6686 case XML_ATTRIBUTE_NODE:
6687 return htmlAttrAllowed(
6688 htmlTagLookup(node->parent->name) , node->name, legacy) ;
6689 default: return HTML_NA ;
6690 }
6691}
6692/************************************************************************
6693 * *
6694 * New set (2.6.0) of simpler and more flexible APIs *
6695 * *
6696 ************************************************************************/
6697/**
6698 * DICT_FREE:
6699 * @str: a string
6700 *
6701 * Free a string if it is not owned by the "dict" dictionary in the
6702 * current scope
6703 */
6704#define DICT_FREE(str) \
6705 if ((str) && ((!dict) || \
6706 (xmlDictOwns(dict, (const xmlChar *)(str)) == 0))) \
6707 xmlFree((char *)(str));
6708
6709/**
6710 * htmlCtxtReset:
6711 * @ctxt: an HTML parser context
6712 *
6713 * Reset a parser context
6714 */
6715void
6716htmlCtxtReset(htmlParserCtxtPtr ctxt)
6717{
6718 xmlParserInputPtr input;
6719 xmlDictPtr dict;
6720
6721 if (ctxt == NULL)
6722 return;
6723
6724 xmlInitParser();
6725 dict = ctxt->dict;
6726
6727 while ((input = inputPop(ctxt)) != NULL) { /* Non consuming */
6728 xmlFreeInputStream(input);
6729 }
6730 ctxt->inputNr = 0;
6731 ctxt->input = NULL;
6732
6733 ctxt->spaceNr = 0;
6734 if (ctxt->spaceTab != NULL) {
6735 ctxt->spaceTab[0] = -1;
6736 ctxt->space = &ctxt->spaceTab[0];
6737 } else {
6738 ctxt->space = NULL;
6739 }
6740
6741
6742 ctxt->nodeNr = 0;
6743 ctxt->node = NULL;
6744
6745 ctxt->nameNr = 0;
6746 ctxt->name = NULL;
6747
6748 DICT_FREE(ctxt->version);
6749 ctxt->version = NULL;
6750 DICT_FREE(ctxt->encoding);
6751 ctxt->encoding = NULL;
6752 DICT_FREE(ctxt->directory);
6753 ctxt->directory = NULL;
6754 DICT_FREE(ctxt->extSubURI);
6755 ctxt->extSubURI = NULL;
6756 DICT_FREE(ctxt->extSubSystem);
6757 ctxt->extSubSystem = NULL;
6758 if (ctxt->myDoc != NULL)
6759 xmlFreeDoc(ctxt->myDoc);
6760 ctxt->myDoc = NULL;
6761
6762 ctxt->standalone = -1;
6763 ctxt->hasExternalSubset = 0;
6764 ctxt->hasPErefs = 0;
6765 ctxt->html = 1;
6766 ctxt->external = 0;
6767 ctxt->instate = XML_PARSER_START;
6768 ctxt->token = 0;
6769
6770 ctxt->wellFormed = 1;
6771 ctxt->nsWellFormed = 1;
6772 ctxt->disableSAX = 0;
6773 ctxt->valid = 1;
6774 ctxt->vctxt.userData = ctxt;
6775 ctxt->vctxt.error = xmlParserValidityError;
6776 ctxt->vctxt.warning = xmlParserValidityWarning;
6777 ctxt->record_info = 0;
6778 ctxt->checkIndex = 0;
6779 ctxt->inSubset = 0;
6780 ctxt->errNo = XML_ERR_OK;
6781 ctxt->depth = 0;
6782 ctxt->charset = XML_CHAR_ENCODING_NONE;
6783 ctxt->catalogs = NULL;
6784 xmlInitNodeInfoSeq(&ctxt->node_seq);
6785
6786 if (ctxt->attsDefault != NULL) {
6787 xmlHashFree(ctxt->attsDefault, xmlHashDefaultDeallocator);
6788 ctxt->attsDefault = NULL;
6789 }
6790 if (ctxt->attsSpecial != NULL) {
6791 xmlHashFree(ctxt->attsSpecial, NULL);
6792 ctxt->attsSpecial = NULL;
6793 }
6794}
6795
6796/**
6797 * htmlCtxtUseOptions:
6798 * @ctxt: an HTML parser context
6799 * @options: a combination of htmlParserOption(s)
6800 *
6801 * Applies the options to the parser context
6802 *
6803 * Returns 0 in case of success, the set of unknown or unimplemented options
6804 * in case of error.
6805 */
6806int
6807htmlCtxtUseOptions(htmlParserCtxtPtr ctxt, int options)
6808{
6809 if (ctxt == NULL)
6810 return(-1);
6811
6812 if (options & HTML_PARSE_NOWARNING) {
6813 ctxt->sax->warning = NULL;
6814 ctxt->vctxt.warning = NULL;
6815 options -= XML_PARSE_NOWARNING;
6816 ctxt->options |= XML_PARSE_NOWARNING;
6817 }
6818 if (options & HTML_PARSE_NOERROR) {
6819 ctxt->sax->error = NULL;
6820 ctxt->vctxt.error = NULL;
6821 ctxt->sax->fatalError = NULL;
6822 options -= XML_PARSE_NOERROR;
6823 ctxt->options |= XML_PARSE_NOERROR;
6824 }
6825 if (options & HTML_PARSE_PEDANTIC) {
6826 ctxt->pedantic = 1;
6827 options -= XML_PARSE_PEDANTIC;
6828 ctxt->options |= XML_PARSE_PEDANTIC;
6829 } else
6830 ctxt->pedantic = 0;
6831 if (options & XML_PARSE_NOBLANKS) {
6832 ctxt->keepBlanks = 0;
6833 ctxt->sax->ignorableWhitespace = xmlSAX2IgnorableWhitespace;
6834 options -= XML_PARSE_NOBLANKS;
6835 ctxt->options |= XML_PARSE_NOBLANKS;
6836 } else
6837 ctxt->keepBlanks = 1;
6838 if (options & HTML_PARSE_RECOVER) {
6839 ctxt->recovery = 1;
6840 options -= HTML_PARSE_RECOVER;
6841 } else
6842 ctxt->recovery = 0;
6843 if (options & HTML_PARSE_COMPACT) {
6844 ctxt->options |= HTML_PARSE_COMPACT;
6845 options -= HTML_PARSE_COMPACT;
6846 }
6847 if (options & XML_PARSE_HUGE) {
6848 ctxt->options |= XML_PARSE_HUGE;
6849 options -= XML_PARSE_HUGE;
6850 }
6851 if (options & HTML_PARSE_NODEFDTD) {
6852 ctxt->options |= HTML_PARSE_NODEFDTD;
6853 options -= HTML_PARSE_NODEFDTD;
6854 }
6855 if (options & HTML_PARSE_IGNORE_ENC) {
6856 ctxt->options |= HTML_PARSE_IGNORE_ENC;
6857 options -= HTML_PARSE_IGNORE_ENC;
6858 }
6859 if (options & HTML_PARSE_NOIMPLIED) {
6860 ctxt->options |= HTML_PARSE_NOIMPLIED;
6861 options -= HTML_PARSE_NOIMPLIED;
6862 }
6863 ctxt->dictNames = 0;
6864 return (options);
6865}
6866
6867/**
6868 * htmlDoRead:
6869 * @ctxt: an HTML parser context
6870 * @URL: the base URL to use for the document
6871 * @encoding: the document encoding, or NULL
6872 * @options: a combination of htmlParserOption(s)
6873 * @reuse: keep the context for reuse
6874 *
6875 * Common front-end for the htmlRead functions
6876 *
6877 * Returns the resulting document tree or NULL
6878 */
6879static htmlDocPtr
6880htmlDoRead(htmlParserCtxtPtr ctxt, const char *URL, const char *encoding,
6881 int options, int reuse)
6882{
6883 htmlDocPtr ret;
6884
6885 htmlCtxtUseOptions(ctxt, options);
6886 ctxt->html = 1;
6887 if (encoding != NULL) {
6888 xmlCharEncodingHandlerPtr hdlr;
6889
6890 hdlr = xmlFindCharEncodingHandler(encoding);
6891 if (hdlr != NULL) {
6892 xmlSwitchToEncoding(ctxt, hdlr);
6893 if (ctxt->input->encoding != NULL)
6894 xmlFree((xmlChar *) ctxt->input->encoding);
6895 ctxt->input->encoding = xmlStrdup((xmlChar *)encoding);
6896 }
6897 }
6898 if ((URL != NULL) && (ctxt->input != NULL) &&
6899 (ctxt->input->filename == NULL))
6900 ctxt->input->filename = (char *) xmlStrdup((const xmlChar *) URL);
6901 htmlParseDocument(ctxt);
6902 ret = ctxt->myDoc;
6903 ctxt->myDoc = NULL;
6904 if (!reuse) {
6905 if ((ctxt->dictNames) &&
6906 (ret != NULL) &&
6907 (ret->dict == ctxt->dict))
6908 ctxt->dict = NULL;
6909 xmlFreeParserCtxt(ctxt);
6910 }
6911 return (ret);
6912}
6913
6914/**
6915 * htmlReadDoc:
6916 * @cur: a pointer to a zero terminated string
6917 * @URL: the base URL to use for the document
6918 * @encoding: the document encoding, or NULL
6919 * @options: a combination of htmlParserOption(s)
6920 *
6921 * parse an XML in-memory document and build a tree.
6922 *
6923 * Returns the resulting document tree
6924 */
6925htmlDocPtr
6926htmlReadDoc(const xmlChar * cur, const char *URL, const char *encoding, int options)
6927{
6928 htmlParserCtxtPtr ctxt;
6929
6930 if (cur == NULL)
6931 return (NULL);
6932
6933 xmlInitParser();
6934 ctxt = htmlCreateDocParserCtxt(cur, NULL);
6935 if (ctxt == NULL)
6936 return (NULL);
6937 return (htmlDoRead(ctxt, URL, encoding, options, 0));
6938}
6939
6940/**
6941 * htmlReadFile:
6942 * @filename: a file or URL
6943 * @encoding: the document encoding, or NULL
6944 * @options: a combination of htmlParserOption(s)
6945 *
6946 * parse an XML file from the filesystem or the network.
6947 *
6948 * Returns the resulting document tree
6949 */
6950htmlDocPtr
6951htmlReadFile(const char *filename, const char *encoding, int options)
6952{
6953 htmlParserCtxtPtr ctxt;
6954
6955 xmlInitParser();
6956 ctxt = htmlCreateFileParserCtxt(filename, encoding);
6957 if (ctxt == NULL)
6958 return (NULL);
6959 return (htmlDoRead(ctxt, NULL, NULL, options, 0));
6960}
6961
6962/**
6963 * htmlReadMemory:
6964 * @buffer: a pointer to a char array
6965 * @size: the size of the array
6966 * @URL: the base URL to use for the document
6967 * @encoding: the document encoding, or NULL
6968 * @options: a combination of htmlParserOption(s)
6969 *
6970 * parse an XML in-memory document and build a tree.
6971 *
6972 * Returns the resulting document tree
6973 */
6974htmlDocPtr
6975htmlReadMemory(const char *buffer, int size, const char *URL, const char *encoding, int options)
6976{
6977 htmlParserCtxtPtr ctxt;
6978
6979 xmlInitParser();
6980 ctxt = xmlCreateMemoryParserCtxt(buffer, size);
6981 if (ctxt == NULL)
6982 return (NULL);
6983 htmlDefaultSAXHandlerInit();
6984 if (ctxt->sax != NULL)
6985 memcpy(ctxt->sax, &htmlDefaultSAXHandler, sizeof(xmlSAXHandlerV1));
6986 return (htmlDoRead(ctxt, URL, encoding, options, 0));
6987}
6988
6989/**
6990 * htmlReadFd:
6991 * @fd: an open file descriptor
6992 * @URL: the base URL to use for the document
6993 * @encoding: the document encoding, or NULL
6994 * @options: a combination of htmlParserOption(s)
6995 *
6996 * parse an HTML from a file descriptor and build a tree.
6997 * NOTE that the file descriptor will not be closed when the
6998 * reader is closed or reset.
6999 *
7000 * Returns the resulting document tree
7001 */
7002htmlDocPtr
7003htmlReadFd(int fd, const char *URL, const char *encoding, int options)
7004{
7005 htmlParserCtxtPtr ctxt;
7006 xmlParserInputBufferPtr input;
7007 htmlParserInputPtr stream;
7008
7009 if (fd < 0)
7010 return (NULL);
7011
7012 xmlInitParser();
7013 input = xmlParserInputBufferCreateFd(fd, XML_CHAR_ENCODING_NONE);
7014 if (input == NULL)
7015 return (NULL);
7016 input->closecallback = NULL;
7017 ctxt = htmlNewParserCtxt();
7018 if (ctxt == NULL) {
7019 xmlFreeParserInputBuffer(input);
7020 return (NULL);
7021 }
7022 stream = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE);
7023 if (stream == NULL) {
7024 xmlFreeParserInputBuffer(input);
7025 htmlFreeParserCtxt(ctxt);
7026 return (NULL);
7027 }
7028 inputPush(ctxt, stream);
7029 return (htmlDoRead(ctxt, URL, encoding, options, 0));
7030}
7031
7032/**
7033 * htmlReadIO:
7034 * @ioread: an I/O read function
7035 * @ioclose: an I/O close function
7036 * @ioctx: an I/O handler
7037 * @URL: the base URL to use for the document
7038 * @encoding: the document encoding, or NULL
7039 * @options: a combination of htmlParserOption(s)
7040 *
7041 * parse an HTML document from I/O functions and source and build a tree.
7042 *
7043 * Returns the resulting document tree
7044 */
7045htmlDocPtr
7046htmlReadIO(xmlInputReadCallback ioread, xmlInputCloseCallback ioclose,
7047 void *ioctx, const char *URL, const char *encoding, int options)
7048{
7049 htmlParserCtxtPtr ctxt;
7050 xmlParserInputBufferPtr input;
7051 xmlParserInputPtr stream;
7052
7053 if (ioread == NULL)
7054 return (NULL);
7055 xmlInitParser();
7056
7057 input = xmlParserInputBufferCreateIO(ioread, ioclose, ioctx,
7058 XML_CHAR_ENCODING_NONE);
7059 if (input == NULL) {
7060 if (ioclose != NULL)
7061 ioclose(ioctx);
7062 return (NULL);
7063 }
7064 ctxt = htmlNewParserCtxt();
7065 if (ctxt == NULL) {
7066 xmlFreeParserInputBuffer(input);
7067 return (NULL);
7068 }
7069 stream = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE);
7070 if (stream == NULL) {
7071 xmlFreeParserInputBuffer(input);
7072 xmlFreeParserCtxt(ctxt);
7073 return (NULL);
7074 }
7075 inputPush(ctxt, stream);
7076 return (htmlDoRead(ctxt, URL, encoding, options, 0));
7077}
7078
7079/**
7080 * htmlCtxtReadDoc:
7081 * @ctxt: an HTML parser context
7082 * @cur: a pointer to a zero terminated string
7083 * @URL: the base URL to use for the document
7084 * @encoding: the document encoding, or NULL
7085 * @options: a combination of htmlParserOption(s)
7086 *
7087 * parse an XML in-memory document and build a tree.
7088 * This reuses the existing @ctxt parser context
7089 *
7090 * Returns the resulting document tree
7091 */
7092htmlDocPtr
7093htmlCtxtReadDoc(htmlParserCtxtPtr ctxt, const xmlChar * cur,
7094 const char *URL, const char *encoding, int options)
7095{
7096 xmlParserInputPtr stream;
7097
7098 if (cur == NULL)
7099 return (NULL);
7100 if (ctxt == NULL)
7101 return (NULL);
7102 xmlInitParser();
7103
7104 htmlCtxtReset(ctxt);
7105
7106 stream = xmlNewStringInputStream(ctxt, cur);
7107 if (stream == NULL) {
7108 return (NULL);
7109 }
7110 inputPush(ctxt, stream);
7111 return (htmlDoRead(ctxt, URL, encoding, options, 1));
7112}
7113
7114/**
7115 * htmlCtxtReadFile:
7116 * @ctxt: an HTML parser context
7117 * @filename: a file or URL
7118 * @encoding: the document encoding, or NULL
7119 * @options: a combination of htmlParserOption(s)
7120 *
7121 * parse an XML file from the filesystem or the network.
7122 * This reuses the existing @ctxt parser context
7123 *
7124 * Returns the resulting document tree
7125 */
7126htmlDocPtr
7127htmlCtxtReadFile(htmlParserCtxtPtr ctxt, const char *filename,
7128 const char *encoding, int options)
7129{
7130 xmlParserInputPtr stream;
7131
7132 if (filename == NULL)
7133 return (NULL);
7134 if (ctxt == NULL)
7135 return (NULL);
7136 xmlInitParser();
7137
7138 htmlCtxtReset(ctxt);
7139
7140 stream = xmlLoadExternalEntity(filename, NULL, ctxt);
7141 if (stream == NULL) {
7142 return (NULL);
7143 }
7144 inputPush(ctxt, stream);
7145 return (htmlDoRead(ctxt, NULL, encoding, options, 1));
7146}
7147
7148/**
7149 * htmlCtxtReadMemory:
7150 * @ctxt: an HTML parser context
7151 * @buffer: a pointer to a char array
7152 * @size: the size of the array
7153 * @URL: the base URL to use for the document
7154 * @encoding: the document encoding, or NULL
7155 * @options: a combination of htmlParserOption(s)
7156 *
7157 * parse an XML in-memory document and build a tree.
7158 * This reuses the existing @ctxt parser context
7159 *
7160 * Returns the resulting document tree
7161 */
7162htmlDocPtr
7163htmlCtxtReadMemory(htmlParserCtxtPtr ctxt, const char *buffer, int size,
7164 const char *URL, const char *encoding, int options)
7165{
7166 xmlParserInputBufferPtr input;
7167 xmlParserInputPtr stream;
7168
7169 if (ctxt == NULL)
7170 return (NULL);
7171 if (buffer == NULL)
7172 return (NULL);
7173 xmlInitParser();
7174
7175 htmlCtxtReset(ctxt);
7176
7177 input = xmlParserInputBufferCreateMem(buffer, size, XML_CHAR_ENCODING_NONE);
7178 if (input == NULL) {
7179 return(NULL);
7180 }
7181
7182 stream = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE);
7183 if (stream == NULL) {
7184 xmlFreeParserInputBuffer(input);
7185 return(NULL);
7186 }
7187
7188 inputPush(ctxt, stream);
7189 return (htmlDoRead(ctxt, URL, encoding, options, 1));
7190}
7191
7192/**
7193 * htmlCtxtReadFd:
7194 * @ctxt: an HTML parser context
7195 * @fd: an open file descriptor
7196 * @URL: the base URL to use for the document
7197 * @encoding: the document encoding, or NULL
7198 * @options: a combination of htmlParserOption(s)
7199 *
7200 * parse an XML from a file descriptor and build a tree.
7201 * This reuses the existing @ctxt parser context
7202 *
7203 * Returns the resulting document tree
7204 */
7205htmlDocPtr
7206htmlCtxtReadFd(htmlParserCtxtPtr ctxt, int fd,
7207 const char *URL, const char *encoding, int options)
7208{
7209 xmlParserInputBufferPtr input;
7210 xmlParserInputPtr stream;
7211
7212 if (fd < 0)
7213 return (NULL);
7214 if (ctxt == NULL)
7215 return (NULL);
7216 xmlInitParser();
7217
7218 htmlCtxtReset(ctxt);
7219
7220
7221 input = xmlParserInputBufferCreateFd(fd, XML_CHAR_ENCODING_NONE);
7222 if (input == NULL)
7223 return (NULL);
7224 stream = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE);
7225 if (stream == NULL) {
7226 xmlFreeParserInputBuffer(input);
7227 return (NULL);
7228 }
7229 inputPush(ctxt, stream);
7230 return (htmlDoRead(ctxt, URL, encoding, options, 1));
7231}
7232
7233/**
7234 * htmlCtxtReadIO:
7235 * @ctxt: an HTML parser context
7236 * @ioread: an I/O read function
7237 * @ioclose: an I/O close function
7238 * @ioctx: an I/O handler
7239 * @URL: the base URL to use for the document
7240 * @encoding: the document encoding, or NULL
7241 * @options: a combination of htmlParserOption(s)
7242 *
7243 * parse an HTML document from I/O functions and source and build a tree.
7244 * This reuses the existing @ctxt parser context
7245 *
7246 * Returns the resulting document tree
7247 */
7248htmlDocPtr
7249htmlCtxtReadIO(htmlParserCtxtPtr ctxt, xmlInputReadCallback ioread,
7250 xmlInputCloseCallback ioclose, void *ioctx,
7251 const char *URL,
7252 const char *encoding, int options)
7253{
7254 xmlParserInputBufferPtr input;
7255 xmlParserInputPtr stream;
7256
7257 if (ioread == NULL)
7258 return (NULL);
7259 if (ctxt == NULL)
7260 return (NULL);
7261 xmlInitParser();
7262
7263 htmlCtxtReset(ctxt);
7264
7265 input = xmlParserInputBufferCreateIO(ioread, ioclose, ioctx,
7266 XML_CHAR_ENCODING_NONE);
7267 if (input == NULL) {
7268 if (ioclose != NULL)
7269 ioclose(ioctx);
7270 return (NULL);
7271 }
7272 stream = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE);
7273 if (stream == NULL) {
7274 xmlFreeParserInputBuffer(input);
7275 return (NULL);
7276 }
7277 inputPush(ctxt, stream);
7278 return (htmlDoRead(ctxt, URL, encoding, options, 1));
7279}
7280
7281#define bottom_HTMLparser
7282#include "elfgcchack.h"
7283#endif /* LIBXML_HTML_ENABLED */
Note: See TracBrowser for help on using the repository browser.

© 2024 Oracle Support Privacy / Do Not Sell My Info Terms of Use Trademark Policy Automated Access Etiquette