VirtualBox

source: vbox/trunk/src/libs/libxml2-2.6.30/HTMLparser.c@ 25275

Last change on this file since 25275 was 6076, checked in by vboxsync, 17 years ago

Merged dmik/s2 branch (r25959:26751) to the trunk.
  • Property svn:eol-style set to native
  • Property svn:keywords set to Date Revision Author Id
File size: 183.3 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 * daniel@veillard.com
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 HAVE_ZLIB_H
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#define HTML_MAX_NAMELEN 1000
48#define HTML_PARSER_BIG_BUFFER_SIZE 1000
49#define HTML_PARSER_BUFFER_SIZE 100
50
51/* #define DEBUG */
52/* #define DEBUG_PUSH */
53
54static int htmlOmittedDefaultValue = 1;
55
56xmlChar * htmlDecodeEntities(htmlParserCtxtPtr ctxt, int len,
57 xmlChar end, xmlChar end2, xmlChar end3);
58static void htmlParseComment(htmlParserCtxtPtr ctxt);
59
60/************************************************************************
61 * *
62 * Some factorized error routines *
63 * *
64 ************************************************************************/
65
66/**
67 * htmlErrMemory:
68 * @ctxt: an HTML parser context
69 * @extra: extra informations
70 *
71 * Handle a redefinition of attribute error
72 */
73static void
74htmlErrMemory(xmlParserCtxtPtr ctxt, const char *extra)
75{
76 if ((ctxt != NULL) && (ctxt->disableSAX != 0) &&
77 (ctxt->instate == XML_PARSER_EOF))
78 return;
79 if (ctxt != NULL) {
80 ctxt->errNo = XML_ERR_NO_MEMORY;
81 ctxt->instate = XML_PARSER_EOF;
82 ctxt->disableSAX = 1;
83 }
84 if (extra)
85 __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER,
86 XML_ERR_NO_MEMORY, XML_ERR_FATAL, NULL, 0, extra,
87 NULL, NULL, 0, 0,
88 "Memory allocation failed : %s\n", extra);
89 else
90 __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER,
91 XML_ERR_NO_MEMORY, XML_ERR_FATAL, NULL, 0, NULL,
92 NULL, NULL, 0, 0, "Memory allocation failed\n");
93}
94
95/**
96 * htmlParseErr:
97 * @ctxt: an HTML parser context
98 * @error: the error number
99 * @msg: the error message
100 * @str1: string infor
101 * @str2: string infor
102 *
103 * Handle a fatal parser error, i.e. violating Well-Formedness constraints
104 */
105static void
106htmlParseErr(xmlParserCtxtPtr ctxt, xmlParserErrors error,
107 const char *msg, const xmlChar *str1, const xmlChar *str2)
108{
109 if ((ctxt != NULL) && (ctxt->disableSAX != 0) &&
110 (ctxt->instate == XML_PARSER_EOF))
111 return;
112 if (ctxt != NULL)
113 ctxt->errNo = error;
114 __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_HTML, error,
115 XML_ERR_ERROR, NULL, 0,
116 (const char *) str1, (const char *) str2,
117 NULL, 0, 0,
118 msg, str1, str2);
119 if (ctxt != NULL)
120 ctxt->wellFormed = 0;
121}
122
123/**
124 * htmlParseErrInt:
125 * @ctxt: an HTML parser context
126 * @error: the error number
127 * @msg: the error message
128 * @val: integer info
129 *
130 * Handle a fatal parser error, i.e. violating Well-Formedness constraints
131 */
132static void
133htmlParseErrInt(xmlParserCtxtPtr ctxt, xmlParserErrors error,
134 const char *msg, int val)
135{
136 if ((ctxt != NULL) && (ctxt->disableSAX != 0) &&
137 (ctxt->instate == XML_PARSER_EOF))
138 return;
139 if (ctxt != NULL)
140 ctxt->errNo = error;
141 __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_HTML, error,
142 XML_ERR_ERROR, NULL, 0, NULL, NULL,
143 NULL, val, 0, msg, val);
144 if (ctxt != NULL)
145 ctxt->wellFormed = 0;
146}
147
148/************************************************************************
149 * *
150 * Parser stacks related functions and macros *
151 * *
152 ************************************************************************/
153
154/**
155 * htmlnamePush:
156 * @ctxt: an HTML parser context
157 * @value: the element name
158 *
159 * Pushes a new element name on top of the name stack
160 *
161 * Returns 0 in case of error, the index in the stack otherwise
162 */
163static int
164htmlnamePush(htmlParserCtxtPtr ctxt, const xmlChar * value)
165{
166 if (ctxt->nameNr >= ctxt->nameMax) {
167 ctxt->nameMax *= 2;
168 ctxt->nameTab = (const xmlChar * *)
169 xmlRealloc((xmlChar * *)ctxt->nameTab,
170 ctxt->nameMax *
171 sizeof(ctxt->nameTab[0]));
172 if (ctxt->nameTab == NULL) {
173 htmlErrMemory(ctxt, NULL);
174 return (0);
175 }
176 }
177 ctxt->nameTab[ctxt->nameNr] = value;
178 ctxt->name = value;
179 return (ctxt->nameNr++);
180}
181/**
182 * htmlnamePop:
183 * @ctxt: an HTML parser context
184 *
185 * Pops the top element name from the name stack
186 *
187 * Returns the name just removed
188 */
189static const xmlChar *
190htmlnamePop(htmlParserCtxtPtr ctxt)
191{
192 const xmlChar *ret;
193
194 if (ctxt->nameNr <= 0)
195 return (NULL);
196 ctxt->nameNr--;
197 if (ctxt->nameNr < 0)
198 return (NULL);
199 if (ctxt->nameNr > 0)
200 ctxt->name = ctxt->nameTab[ctxt->nameNr - 1];
201 else
202 ctxt->name = NULL;
203 ret = ctxt->nameTab[ctxt->nameNr];
204 ctxt->nameTab[ctxt->nameNr] = NULL;
205 return (ret);
206}
207
208/*
209 * Macros for accessing the content. Those should be used only by the parser,
210 * and not exported.
211 *
212 * Dirty macros, i.e. one need to make assumption on the context to use them
213 *
214 * CUR_PTR return the current pointer to the xmlChar to be parsed.
215 * CUR returns the current xmlChar value, i.e. a 8 bit value if compiled
216 * in ISO-Latin or UTF-8, and the current 16 bit value if compiled
217 * in UNICODE mode. This should be used internally by the parser
218 * only to compare to ASCII values otherwise it would break when
219 * running with UTF-8 encoding.
220 * NXT(n) returns the n'th next xmlChar. Same as CUR is should be used only
221 * to compare on ASCII based substring.
222 * UPP(n) returns the n'th next xmlChar converted to uppercase. Same as CUR
223 * it should be used only to compare on ASCII based substring.
224 * SKIP(n) Skip n xmlChar, and must also be used only to skip ASCII defined
225 * strings without newlines within the parser.
226 *
227 * Clean macros, not dependent of an ASCII context, expect UTF-8 encoding
228 *
229 * CURRENT Returns the current char value, with the full decoding of
230 * UTF-8 if we are using this mode. It returns an int.
231 * NEXT Skip to the next character, this does the proper decoding
232 * in UTF-8 mode. It also pop-up unfinished entities on the fly.
233 * NEXTL(l) Skip the current unicode character of l xmlChars long.
234 * COPY(to) copy one char to *to, increment CUR_PTR and to accordingly
235 */
236
237#define UPPER (toupper(*ctxt->input->cur))
238
239#define SKIP(val) ctxt->nbChars += (val),ctxt->input->cur += (val),ctxt->input->col+=(val)
240
241#define NXT(val) ctxt->input->cur[(val)]
242
243#define UPP(val) (toupper(ctxt->input->cur[(val)]))
244
245#define CUR_PTR ctxt->input->cur
246
247#define SHRINK if ((ctxt->input->cur - ctxt->input->base > 2 * INPUT_CHUNK) && \
248 (ctxt->input->end - ctxt->input->cur < 2 * INPUT_CHUNK)) \
249 xmlParserInputShrink(ctxt->input)
250
251#define GROW if ((ctxt->progressive == 0) && \
252 (ctxt->input->end - ctxt->input->cur < INPUT_CHUNK)) \
253 xmlParserInputGrow(ctxt->input, INPUT_CHUNK)
254
255#define CURRENT ((int) (*ctxt->input->cur))
256
257#define SKIP_BLANKS htmlSkipBlankChars(ctxt)
258
259/* Inported from XML */
260
261/* #define CUR (ctxt->token ? ctxt->token : (int) (*ctxt->input->cur)) */
262#define CUR ((int) (*ctxt->input->cur))
263#define NEXT xmlNextChar(ctxt)
264
265#define RAW (ctxt->token ? -1 : (*ctxt->input->cur))
266#define NXT(val) ctxt->input->cur[(val)]
267#define CUR_PTR ctxt->input->cur
268
269
270#define NEXTL(l) do { \
271 if (*(ctxt->input->cur) == '\n') { \
272 ctxt->input->line++; ctxt->input->col = 1; \
273 } else ctxt->input->col++; \
274 ctxt->token = 0; ctxt->input->cur += l; ctxt->nbChars++; \
275 } while (0)
276
277/************
278 \
279 if (*ctxt->input->cur == '%') xmlParserHandlePEReference(ctxt); \
280 if (*ctxt->input->cur == '&') xmlParserHandleReference(ctxt);
281 ************/
282
283#define CUR_CHAR(l) htmlCurrentChar(ctxt, &l)
284#define CUR_SCHAR(s, l) xmlStringCurrentChar(ctxt, s, &l)
285
286#define COPY_BUF(l,b,i,v) \
287 if (l == 1) b[i++] = (xmlChar) v; \
288 else i += xmlCopyChar(l,&b[i],v)
289
290/**
291 * htmlCurrentChar:
292 * @ctxt: the HTML parser context
293 * @len: pointer to the length of the char read
294 *
295 * The current char value, if using UTF-8 this may actually span multiple
296 * bytes in the input buffer. Implement the end of line normalization:
297 * 2.11 End-of-Line Handling
298 * If the encoding is unspecified, in the case we find an ISO-Latin-1
299 * char, then the encoding converter is plugged in automatically.
300 *
301 * Returns the current char value and its length
302 */
303
304static int
305htmlCurrentChar(xmlParserCtxtPtr ctxt, int *len) {
306 if (ctxt->instate == XML_PARSER_EOF)
307 return(0);
308
309 if (ctxt->token != 0) {
310 *len = 0;
311 return(ctxt->token);
312 }
313 if (ctxt->charset == XML_CHAR_ENCODING_UTF8) {
314 /*
315 * We are supposed to handle UTF8, check it's valid
316 * From rfc2044: encoding of the Unicode values on UTF-8:
317 *
318 * UCS-4 range (hex.) UTF-8 octet sequence (binary)
319 * 0000 0000-0000 007F 0xxxxxxx
320 * 0000 0080-0000 07FF 110xxxxx 10xxxxxx
321 * 0000 0800-0000 FFFF 1110xxxx 10xxxxxx 10xxxxxx
322 *
323 * Check for the 0x110000 limit too
324 */
325 const unsigned char *cur = ctxt->input->cur;
326 unsigned char c;
327 unsigned int val;
328
329 c = *cur;
330 if (c & 0x80) {
331 if (cur[1] == 0)
332 xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
333 if ((cur[1] & 0xc0) != 0x80)
334 goto encoding_error;
335 if ((c & 0xe0) == 0xe0) {
336
337 if (cur[2] == 0)
338 xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
339 if ((cur[2] & 0xc0) != 0x80)
340 goto encoding_error;
341 if ((c & 0xf0) == 0xf0) {
342 if (cur[3] == 0)
343 xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
344 if (((c & 0xf8) != 0xf0) ||
345 ((cur[3] & 0xc0) != 0x80))
346 goto encoding_error;
347 /* 4-byte code */
348 *len = 4;
349 val = (cur[0] & 0x7) << 18;
350 val |= (cur[1] & 0x3f) << 12;
351 val |= (cur[2] & 0x3f) << 6;
352 val |= cur[3] & 0x3f;
353 } else {
354 /* 3-byte code */
355 *len = 3;
356 val = (cur[0] & 0xf) << 12;
357 val |= (cur[1] & 0x3f) << 6;
358 val |= cur[2] & 0x3f;
359 }
360 } else {
361 /* 2-byte code */
362 *len = 2;
363 val = (cur[0] & 0x1f) << 6;
364 val |= cur[1] & 0x3f;
365 }
366 if (!IS_CHAR(val)) {
367 htmlParseErrInt(ctxt, XML_ERR_INVALID_CHAR,
368 "Char 0x%X out of allowed range\n", val);
369 }
370 return(val);
371 } else {
372 /* 1-byte code */
373 *len = 1;
374 return((int) *ctxt->input->cur);
375 }
376 }
377 /*
378 * Assume it's a fixed length encoding (1) with
379 * a compatible encoding for the ASCII set, since
380 * XML constructs only use < 128 chars
381 */
382 *len = 1;
383 if ((int) *ctxt->input->cur < 0x80)
384 return((int) *ctxt->input->cur);
385
386 /*
387 * Humm this is bad, do an automatic flow conversion
388 */
389 xmlSwitchEncoding(ctxt, XML_CHAR_ENCODING_8859_1);
390 ctxt->charset = XML_CHAR_ENCODING_UTF8;
391 return(xmlCurrentChar(ctxt, len));
392
393encoding_error:
394 /*
395 * If we detect an UTF8 error that probably mean that the
396 * input encoding didn't get properly advertized in the
397 * declaration header. Report the error and switch the encoding
398 * to ISO-Latin-1 (if you don't like this policy, just declare the
399 * encoding !)
400 */
401 {
402 char buffer[150];
403
404 if (ctxt->input->end - ctxt->input->cur >= 4) {
405 snprintf(buffer, 149, "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n",
406 ctxt->input->cur[0], ctxt->input->cur[1],
407 ctxt->input->cur[2], ctxt->input->cur[3]);
408 } else {
409 snprintf(buffer, 149, "Bytes: 0x%02X\n", ctxt->input->cur[0]);
410 }
411 htmlParseErr(ctxt, XML_ERR_INVALID_ENCODING,
412 "Input is not proper UTF-8, indicate encoding !\n",
413 BAD_CAST buffer, NULL);
414 }
415
416 ctxt->charset = XML_CHAR_ENCODING_8859_1;
417 *len = 1;
418 return((int) *ctxt->input->cur);
419}
420
421/**
422 * htmlSkipBlankChars:
423 * @ctxt: the HTML parser context
424 *
425 * skip all blanks character found at that point in the input streams.
426 *
427 * Returns the number of space chars skipped
428 */
429
430static int
431htmlSkipBlankChars(xmlParserCtxtPtr ctxt) {
432 int res = 0;
433
434 while (IS_BLANK_CH(*(ctxt->input->cur))) {
435 if ((*ctxt->input->cur == 0) &&
436 (xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0)) {
437 xmlPopInput(ctxt);
438 } else {
439 if (*(ctxt->input->cur) == '\n') {
440 ctxt->input->line++; ctxt->input->col = 1;
441 } else ctxt->input->col++;
442 ctxt->input->cur++;
443 ctxt->nbChars++;
444 if (*ctxt->input->cur == 0)
445 xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
446 }
447 res++;
448 }
449 return(res);
450}
451
452
453
454/************************************************************************
455 * *
456 * The list of HTML elements and their properties *
457 * *
458 ************************************************************************/
459
460/*
461 * Start Tag: 1 means the start tag can be ommited
462 * End Tag: 1 means the end tag can be ommited
463 * 2 means it's forbidden (empty elements)
464 * 3 means the tag is stylistic and should be closed easily
465 * Depr: this element is deprecated
466 * DTD: 1 means that this element is valid only in the Loose DTD
467 * 2 means that this element is valid only in the Frameset DTD
468 *
469 * Name,Start Tag,End Tag,Save End,Empty,Deprecated,DTD,inline,Description
470 , subElements , impliedsubelt , Attributes, userdata
471 */
472
473/* Definitions and a couple of vars for HTML Elements */
474
475#define FONTSTYLE "tt", "i", "b", "u", "s", "strike", "big", "small"
476#define NB_FONTSTYLE 8
477#define PHRASE "em", "strong", "dfn", "code", "samp", "kbd", "var", "cite", "abbr", "acronym"
478#define NB_PHRASE 10
479#define SPECIAL "a", "img", "applet", "embed", "object", "font", "basefont", "br", "script", "map", "q", "sub", "sup", "span", "bdo", "iframe"
480#define NB_SPECIAL 16
481#define INLINE PCDATA FONTSTYLE PHRASE SPECIAL FORMCTRL
482#define NB_INLINE NB_PCDATA + NB_FONTSTYLE + NB_PHRASE + NB_SPECIAL + NB_FORMCTRL
483#define BLOCK HEADING, LIST "pre", "p", "dl", "div", "center", "noscript", "noframes", "blockquote", "form", "isindex", "hr", "table", "fieldset", "address"
484#define NB_BLOCK NB_HEADING + NB_LIST + 14
485#define FORMCTRL "input", "select", "textarea", "label", "button"
486#define NB_FORMCTRL 5
487#define PCDATA
488#define NB_PCDATA 0
489#define HEADING "h1", "h2", "h3", "h4", "h5", "h6"
490#define NB_HEADING 6
491#define LIST "ul", "ol", "dir", "menu"
492#define NB_LIST 4
493#define MODIFIER
494#define NB_MODIFIER 0
495#define FLOW BLOCK,INLINE
496#define NB_FLOW NB_BLOCK + NB_INLINE
497#define EMPTY NULL
498
499
500static const char* const html_flow[] = { FLOW, NULL } ;
501static const char* const html_inline[] = { INLINE, NULL } ;
502
503/* placeholders: elts with content but no subelements */
504static const char* const html_pcdata[] = { NULL } ;
505#define html_cdata html_pcdata
506
507
508/* ... and for HTML Attributes */
509
510#define COREATTRS "id", "class", "style", "title"
511#define NB_COREATTRS 4
512#define I18N "lang", "dir"
513#define NB_I18N 2
514#define EVENTS "onclick", "ondblclick", "onmousedown", "onmouseup", "onmouseover", "onmouseout", "onkeypress", "onkeydown", "onkeyup"
515#define NB_EVENTS 9
516#define ATTRS COREATTRS,I18N,EVENTS
517#define NB_ATTRS NB_NB_COREATTRS + NB_I18N + NB_EVENTS
518#define CELLHALIGN "align", "char", "charoff"
519#define NB_CELLHALIGN 3
520#define CELLVALIGN "valign"
521#define NB_CELLVALIGN 1
522
523static const char* const html_attrs[] = { ATTRS, NULL } ;
524static const char* const core_i18n_attrs[] = { COREATTRS, I18N, NULL } ;
525static const char* const core_attrs[] = { COREATTRS, NULL } ;
526static const char* const i18n_attrs[] = { I18N, NULL } ;
527
528
529/* Other declarations that should go inline ... */
530static const char* const a_attrs[] = { ATTRS, "charset", "type", "name",
531 "href", "hreflang", "rel", "rev", "accesskey", "shape", "coords",
532 "tabindex", "onfocus", "onblur", NULL } ;
533static const char* const target_attr[] = { "target", NULL } ;
534static const char* const rows_cols_attr[] = { "rows", "cols", NULL } ;
535static const char* const alt_attr[] = { "alt", NULL } ;
536static const char* const src_alt_attrs[] = { "src", "alt", NULL } ;
537static const char* const href_attrs[] = { "href", NULL } ;
538static const char* const clear_attrs[] = { "clear", NULL } ;
539static const char* const inline_p[] = { INLINE, "p", NULL } ;
540
541static const char* const flow_param[] = { FLOW, "param", NULL } ;
542static const char* const applet_attrs[] = { COREATTRS , "codebase",
543 "archive", "alt", "name", "height", "width", "align",
544 "hspace", "vspace", NULL } ;
545static const char* const area_attrs[] = { "shape", "coords", "href", "nohref",
546 "tabindex", "accesskey", "onfocus", "onblur", NULL } ;
547static const char* const basefont_attrs[] =
548 { "id", "size", "color", "face", NULL } ;
549static const char* const quote_attrs[] = { ATTRS, "cite", NULL } ;
550static const char* const body_contents[] = { FLOW, "ins", "del", NULL } ;
551static const char* const body_attrs[] = { ATTRS, "onload", "onunload", NULL } ;
552static const char* const body_depr[] = { "background", "bgcolor", "text",
553 "link", "vlink", "alink", NULL } ;
554static const char* const button_attrs[] = { ATTRS, "name", "value", "type",
555 "disabled", "tabindex", "accesskey", "onfocus", "onblur", NULL } ;
556
557
558static const char* const col_attrs[] = { ATTRS, "span", "width", CELLHALIGN, CELLVALIGN, NULL } ;
559static const char* const col_elt[] = { "col", NULL } ;
560static const char* const edit_attrs[] = { ATTRS, "datetime", "cite", NULL } ;
561static const char* const compact_attrs[] = { ATTRS, "compact", NULL } ;
562static const char* const dl_contents[] = { "dt", "dd", NULL } ;
563static const char* const compact_attr[] = { "compact", NULL } ;
564static const char* const label_attr[] = { "label", NULL } ;
565static const char* const fieldset_contents[] = { FLOW, "legend" } ;
566static const char* const font_attrs[] = { COREATTRS, I18N, "size", "color", "face" , NULL } ;
567static const char* const form_contents[] = { HEADING, LIST, INLINE, "pre", "p", "div", "center", "noscript", "noframes", "blockquote", "isindex", "hr", "table", "fieldset", "address", NULL } ;
568static const char* const form_attrs[] = { ATTRS, "method", "enctype", "accept", "name", "onsubmit", "onreset", "accept-charset", NULL } ;
569static const char* const frame_attrs[] = { COREATTRS, "longdesc", "name", "src", "frameborder", "marginwidth", "marginheight", "noresize", "scrolling" , NULL } ;
570static const char* const frameset_attrs[] = { COREATTRS, "rows", "cols", "onload", "onunload", NULL } ;
571static const char* const frameset_contents[] = { "frameset", "frame", "noframes", NULL } ;
572static const char* const head_attrs[] = { I18N, "profile", NULL } ;
573static const char* const head_contents[] = { "title", "isindex", "base", "script", "style", "meta", "link", "object", NULL } ;
574static const char* const hr_depr[] = { "align", "noshade", "size", "width", NULL } ;
575static const char* const version_attr[] = { "version", NULL } ;
576static const char* const html_content[] = { "head", "body", "frameset", NULL } ;
577static const char* const iframe_attrs[] = { COREATTRS, "longdesc", "name", "src", "frameborder", "marginwidth", "marginheight", "scrolling", "align", "height", "width", NULL } ;
578static const char* const img_attrs[] = { ATTRS, "longdesc", "name", "height", "width", "usemap", "ismap", NULL } ;
579static 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 } ;
580static 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 } ;
581static const char* const prompt_attrs[] = { COREATTRS, I18N, "prompt", NULL } ;
582static const char* const label_attrs[] = { ATTRS, "for", "accesskey", "onfocus", "onblur", NULL } ;
583static const char* const legend_attrs[] = { ATTRS, "accesskey", NULL } ;
584static const char* const align_attr[] = { "align", NULL } ;
585static const char* const link_attrs[] = { ATTRS, "charset", "href", "hreflang", "type", "rel", "rev", "media", NULL } ;
586static const char* const map_contents[] = { BLOCK, "area", NULL } ;
587static const char* const name_attr[] = { "name", NULL } ;
588static const char* const action_attr[] = { "action", NULL } ;
589static const char* const blockli_elt[] = { BLOCK, "li", NULL } ;
590static const char* const meta_attrs[] = { I18N, "http-equiv", "name", "scheme", NULL } ;
591static const char* const content_attr[] = { "content", NULL } ;
592static const char* const type_attr[] = { "type", NULL } ;
593static const char* const noframes_content[] = { "body", FLOW MODIFIER, NULL } ;
594static const char* const object_contents[] = { FLOW, "param", NULL } ;
595static const char* const object_attrs[] = { ATTRS, "declare", "classid", "codebase", "data", "type", "codetype", "archive", "standby", "height", "width", "usemap", "name", "tabindex", NULL } ;
596static const char* const object_depr[] = { "align", "border", "hspace", "vspace", NULL } ;
597static const char* const ol_attrs[] = { "type", "compact", "start", NULL} ;
598static const char* const option_elt[] = { "option", NULL } ;
599static const char* const optgroup_attrs[] = { ATTRS, "disabled", NULL } ;
600static const char* const option_attrs[] = { ATTRS, "disabled", "label", "selected", "value", NULL } ;
601static const char* const param_attrs[] = { "id", "value", "valuetype", "type", NULL } ;
602static const char* const width_attr[] = { "width", NULL } ;
603static const char* const pre_content[] = { PHRASE, "tt", "i", "b", "u", "s", "strike", "a", "br", "script", "map", "q", "span", "bdo", "iframe", NULL } ;
604static const char* const script_attrs[] = { "charset", "src", "defer", "event", "for", NULL } ;
605static const char* const language_attr[] = { "language", NULL } ;
606static const char* const select_content[] = { "optgroup", "option", NULL } ;
607static const char* const select_attrs[] = { ATTRS, "name", "size", "multiple", "disabled", "tabindex", "onfocus", "onblur", "onchange", NULL } ;
608static const char* const style_attrs[] = { I18N, "media", "title", NULL } ;
609static const char* const table_attrs[] = { ATTRS "summary", "width", "border", "frame", "rules", "cellspacing", "cellpadding", "datapagesize", NULL } ;
610static const char* const table_depr[] = { "align", "bgcolor", NULL } ;
611static const char* const table_contents[] = { "caption", "col", "colgroup", "thead", "tfoot", "tbody", "tr", NULL} ;
612static const char* const tr_elt[] = { "tr", NULL } ;
613static const char* const talign_attrs[] = { ATTRS, CELLHALIGN, CELLVALIGN, NULL} ;
614static const char* const th_td_depr[] = { "nowrap", "bgcolor", "width", "height", NULL } ;
615static const char* const th_td_attr[] = { ATTRS, "abbr", "axis", "headers", "scope", "rowspan", "colspan", CELLHALIGN, CELLVALIGN, NULL } ;
616static const char* const textarea_attrs[] = { ATTRS, "name", "disabled", "readonly", "tabindex", "accesskey", "onfocus", "onblur", "onselect", "onchange", NULL } ;
617static const char* const tr_contents[] = { "th", "td", NULL } ;
618static const char* const bgcolor_attr[] = { "bgcolor", NULL } ;
619static const char* const li_elt[] = { "li", NULL } ;
620static const char* const ul_depr[] = { "type", "compact", NULL} ;
621static const char* const dir_attr[] = { "dir", NULL} ;
622
623#define DECL (const char**)
624
625static const htmlElemDesc
626html40ElementTable[] = {
627{ "a", 0, 0, 0, 0, 0, 0, 1, "anchor ",
628 DECL html_inline , NULL , DECL a_attrs , DECL target_attr, NULL
629},
630{ "abbr", 0, 0, 0, 0, 0, 0, 1, "abbreviated form",
631 DECL html_inline , NULL , DECL html_attrs, NULL, NULL
632},
633{ "acronym", 0, 0, 0, 0, 0, 0, 1, "",
634 DECL html_inline , NULL , DECL html_attrs, NULL, NULL
635},
636{ "address", 0, 0, 0, 0, 0, 0, 0, "information on author ",
637 DECL inline_p , NULL , DECL html_attrs, NULL, NULL
638},
639{ "applet", 0, 0, 0, 0, 1, 1, 2, "java applet ",
640 DECL flow_param , NULL , NULL , DECL applet_attrs, NULL
641},
642{ "area", 0, 2, 2, 1, 0, 0, 0, "client-side image map area ",
643 EMPTY , NULL , DECL area_attrs , DECL target_attr, DECL alt_attr
644},
645{ "b", 0, 3, 0, 0, 0, 0, 1, "bold text style",
646 DECL html_inline , NULL , DECL html_attrs, NULL, NULL
647},
648{ "base", 0, 2, 2, 1, 0, 0, 0, "document base uri ",
649 EMPTY , NULL , NULL , DECL target_attr, DECL href_attrs
650},
651{ "basefont", 0, 2, 2, 1, 1, 1, 1, "base font size " ,
652 EMPTY , NULL , NULL, DECL basefont_attrs, NULL
653},
654{ "bdo", 0, 0, 0, 0, 0, 0, 1, "i18n bidi over-ride ",
655 DECL html_inline , NULL , DECL core_i18n_attrs, NULL, DECL dir_attr
656},
657{ "big", 0, 3, 0, 0, 0, 0, 1, "large text style",
658 DECL html_inline , NULL , DECL html_attrs, NULL, NULL
659},
660{ "blockquote", 0, 0, 0, 0, 0, 0, 0, "long quotation ",
661 DECL html_flow , NULL , DECL quote_attrs , NULL, NULL
662},
663{ "body", 1, 1, 0, 0, 0, 0, 0, "document body ",
664 DECL body_contents , "div" , DECL body_attrs, DECL body_depr, NULL
665},
666{ "br", 0, 2, 2, 1, 0, 0, 1, "forced line break ",
667 EMPTY , NULL , DECL core_attrs, DECL clear_attrs , NULL
668},
669{ "button", 0, 0, 0, 0, 0, 0, 2, "push button ",
670 DECL html_flow MODIFIER , NULL , DECL button_attrs, NULL, NULL
671},
672{ "caption", 0, 0, 0, 0, 0, 0, 0, "table caption ",
673 DECL html_inline , NULL , DECL html_attrs, NULL, NULL
674},
675{ "center", 0, 3, 0, 0, 1, 1, 0, "shorthand for div align=center ",
676 DECL html_flow , NULL , NULL, DECL html_attrs, NULL
677},
678{ "cite", 0, 0, 0, 0, 0, 0, 1, "citation",
679 DECL html_inline , NULL , DECL html_attrs, NULL, NULL
680},
681{ "code", 0, 0, 0, 0, 0, 0, 1, "computer code fragment",
682 DECL html_inline , NULL , DECL html_attrs, NULL, NULL
683},
684{ "col", 0, 2, 2, 1, 0, 0, 0, "table column ",
685 EMPTY , NULL , DECL col_attrs , NULL, NULL
686},
687{ "colgroup", 0, 1, 0, 0, 0, 0, 0, "table column group ",
688 DECL col_elt , "col" , DECL col_attrs , NULL, NULL
689},
690{ "dd", 0, 1, 0, 0, 0, 0, 0, "definition description ",
691 DECL html_flow , NULL , DECL html_attrs, NULL, NULL
692},
693{ "del", 0, 0, 0, 0, 0, 0, 2, "deleted text ",
694 DECL html_flow , NULL , DECL edit_attrs , NULL, NULL
695},
696{ "dfn", 0, 0, 0, 0, 0, 0, 1, "instance definition",
697 DECL html_inline , NULL , DECL html_attrs, NULL, NULL
698},
699{ "dir", 0, 0, 0, 0, 1, 1, 0, "directory list",
700 DECL blockli_elt, "li" , NULL, DECL compact_attrs, NULL
701},
702{ "div", 0, 0, 0, 0, 0, 0, 0, "generic language/style container",
703 DECL html_flow, NULL, DECL html_attrs, DECL align_attr, NULL
704},
705{ "dl", 0, 0, 0, 0, 0, 0, 0, "definition list ",
706 DECL dl_contents , "dd" , DECL html_attrs, DECL compact_attr, NULL
707},
708{ "dt", 0, 1, 0, 0, 0, 0, 0, "definition term ",
709 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
710},
711{ "em", 0, 3, 0, 0, 0, 0, 1, "emphasis",
712 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
713},
714{ "embed", 0, 1, 2, 0, 1, 1, 1, "generic embedded object ",
715 EMPTY, NULL, DECL embed_attrs, NULL, NULL
716},
717{ "fieldset", 0, 0, 0, 0, 0, 0, 0, "form control group ",
718 DECL fieldset_contents , NULL, DECL html_attrs, NULL, NULL
719},
720{ "font", 0, 3, 0, 0, 1, 1, 1, "local change to font ",
721 DECL html_inline, NULL, NULL, DECL font_attrs, NULL
722},
723{ "form", 0, 0, 0, 0, 0, 0, 0, "interactive form ",
724 DECL form_contents, "fieldset", DECL form_attrs , DECL target_attr, DECL action_attr
725},
726{ "frame", 0, 2, 2, 1, 0, 2, 0, "subwindow " ,
727 EMPTY, NULL, NULL, DECL frame_attrs, NULL
728},
729{ "frameset", 0, 0, 0, 0, 0, 2, 0, "window subdivision" ,
730 DECL frameset_contents, "noframes" , NULL , DECL frameset_attrs, NULL
731},
732{ "h1", 0, 0, 0, 0, 0, 0, 0, "heading ",
733 DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
734},
735{ "h2", 0, 0, 0, 0, 0, 0, 0, "heading ",
736 DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
737},
738{ "h3", 0, 0, 0, 0, 0, 0, 0, "heading ",
739 DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
740},
741{ "h4", 0, 0, 0, 0, 0, 0, 0, "heading ",
742 DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
743},
744{ "h5", 0, 0, 0, 0, 0, 0, 0, "heading ",
745 DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
746},
747{ "h6", 0, 0, 0, 0, 0, 0, 0, "heading ",
748 DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
749},
750{ "head", 1, 1, 0, 0, 0, 0, 0, "document head ",
751 DECL head_contents, NULL, DECL head_attrs, NULL, NULL
752},
753{ "hr", 0, 2, 2, 1, 0, 0, 0, "horizontal rule " ,
754 EMPTY, NULL, DECL html_attrs, DECL hr_depr, NULL
755},
756{ "html", 1, 1, 0, 0, 0, 0, 0, "document root element ",
757 DECL html_content , NULL , DECL i18n_attrs, DECL version_attr, NULL
758},
759{ "i", 0, 3, 0, 0, 0, 0, 1, "italic text style",
760 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
761},
762{ "iframe", 0, 0, 0, 0, 0, 1, 2, "inline subwindow ",
763 DECL html_flow, NULL, NULL, DECL iframe_attrs, NULL
764},
765{ "img", 0, 2, 2, 1, 0, 0, 1, "embedded image ",
766 EMPTY, NULL, DECL img_attrs, DECL align_attr, DECL src_alt_attrs
767},
768{ "input", 0, 2, 2, 1, 0, 0, 1, "form control ",
769 EMPTY, NULL, DECL input_attrs , DECL align_attr, NULL
770},
771{ "ins", 0, 0, 0, 0, 0, 0, 2, "inserted text",
772 DECL html_flow, NULL, DECL edit_attrs, NULL, NULL
773},
774{ "isindex", 0, 2, 2, 1, 1, 1, 0, "single line prompt ",
775 EMPTY, NULL, NULL, DECL prompt_attrs, NULL
776},
777{ "kbd", 0, 0, 0, 0, 0, 0, 1, "text to be entered by the user",
778 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
779},
780{ "label", 0, 0, 0, 0, 0, 0, 1, "form field label text ",
781 DECL html_inline MODIFIER, NULL, DECL label_attrs , NULL, NULL
782},
783{ "legend", 0, 0, 0, 0, 0, 0, 0, "fieldset legend ",
784 DECL html_inline, NULL, DECL legend_attrs , DECL align_attr, NULL
785},
786{ "li", 0, 1, 1, 0, 0, 0, 0, "list item ",
787 DECL html_flow, NULL, DECL html_attrs, NULL, NULL
788},
789{ "link", 0, 2, 2, 1, 0, 0, 0, "a media-independent link ",
790 EMPTY, NULL, DECL link_attrs, DECL target_attr, NULL
791},
792{ "map", 0, 0, 0, 0, 0, 0, 2, "client-side image map ",
793 DECL map_contents , NULL, DECL html_attrs , NULL, DECL name_attr
794},
795{ "menu", 0, 0, 0, 0, 1, 1, 0, "menu list ",
796 DECL blockli_elt , NULL, NULL, DECL compact_attrs, NULL
797},
798{ "meta", 0, 2, 2, 1, 0, 0, 0, "generic metainformation ",
799 EMPTY, NULL, DECL meta_attrs , NULL , DECL content_attr
800},
801{ "noframes", 0, 0, 0, 0, 0, 2, 0, "alternate content container for non frame-based rendering ",
802 DECL noframes_content, "body" , DECL html_attrs, NULL, NULL
803},
804{ "noscript", 0, 0, 0, 0, 0, 0, 0, "alternate content container for non script-based rendering ",
805 DECL html_flow, "div", DECL html_attrs, NULL, NULL
806},
807{ "object", 0, 0, 0, 0, 0, 0, 2, "generic embedded object ",
808 DECL object_contents , "div" , DECL object_attrs, DECL object_depr, NULL
809},
810{ "ol", 0, 0, 0, 0, 0, 0, 0, "ordered list ",
811 DECL li_elt , "li" , DECL html_attrs, DECL ol_attrs, NULL
812},
813{ "optgroup", 0, 0, 0, 0, 0, 0, 0, "option group ",
814 DECL option_elt , "option", DECL optgroup_attrs, NULL, DECL label_attr
815},
816{ "option", 0, 1, 0, 0, 0, 0, 0, "selectable choice " ,
817 DECL html_pcdata, NULL, DECL option_attrs, NULL, NULL
818},
819{ "p", 0, 1, 0, 0, 0, 0, 0, "paragraph ",
820 DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
821},
822{ "param", 0, 2, 2, 1, 0, 0, 0, "named property value ",
823 EMPTY, NULL, DECL param_attrs, NULL, DECL name_attr
824},
825{ "pre", 0, 0, 0, 0, 0, 0, 0, "preformatted text ",
826 DECL pre_content, NULL, DECL html_attrs, DECL width_attr, NULL
827},
828{ "q", 0, 0, 0, 0, 0, 0, 1, "short inline quotation ",
829 DECL html_inline, NULL, DECL quote_attrs, NULL, NULL
830},
831{ "s", 0, 3, 0, 0, 1, 1, 1, "strike-through text style",
832 DECL html_inline, NULL, NULL, DECL html_attrs, NULL
833},
834{ "samp", 0, 0, 0, 0, 0, 0, 1, "sample program output, scripts, etc.",
835 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
836},
837{ "script", 0, 0, 0, 0, 0, 0, 2, "script statements ",
838 DECL html_cdata, NULL, DECL script_attrs, DECL language_attr, DECL type_attr
839},
840{ "select", 0, 0, 0, 0, 0, 0, 1, "option selector ",
841 DECL select_content, NULL, DECL select_attrs, NULL, NULL
842},
843{ "small", 0, 3, 0, 0, 0, 0, 1, "small text style",
844 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
845},
846{ "span", 0, 0, 0, 0, 0, 0, 1, "generic language/style container ",
847 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
848},
849{ "strike", 0, 3, 0, 0, 1, 1, 1, "strike-through text",
850 DECL html_inline, NULL, NULL, DECL html_attrs, NULL
851},
852{ "strong", 0, 3, 0, 0, 0, 0, 1, "strong emphasis",
853 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
854},
855{ "style", 0, 0, 0, 0, 0, 0, 0, "style info ",
856 DECL html_cdata, NULL, DECL style_attrs, NULL, DECL type_attr
857},
858{ "sub", 0, 3, 0, 0, 0, 0, 1, "subscript",
859 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
860},
861{ "sup", 0, 3, 0, 0, 0, 0, 1, "superscript ",
862 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
863},
864{ "table", 0, 0, 0, 0, 0, 0, 0, "",
865 DECL table_contents , "tr" , DECL table_attrs , DECL table_depr, NULL
866},
867{ "tbody", 1, 0, 0, 0, 0, 0, 0, "table body ",
868 DECL tr_elt , "tr" , DECL talign_attrs, NULL, NULL
869},
870{ "td", 0, 0, 0, 0, 0, 0, 0, "table data cell",
871 DECL html_flow, NULL, DECL th_td_attr, DECL th_td_depr, NULL
872},
873{ "textarea", 0, 0, 0, 0, 0, 0, 1, "multi-line text field ",
874 DECL html_pcdata, NULL, DECL textarea_attrs, NULL, DECL rows_cols_attr
875},
876{ "tfoot", 0, 1, 0, 0, 0, 0, 0, "table footer ",
877 DECL tr_elt , "tr" , DECL talign_attrs, NULL, NULL
878},
879{ "th", 0, 1, 0, 0, 0, 0, 0, "table header cell",
880 DECL html_flow, NULL, DECL th_td_attr, DECL th_td_depr, NULL
881},
882{ "thead", 0, 1, 0, 0, 0, 0, 0, "table header ",
883 DECL tr_elt , "tr" , DECL talign_attrs, NULL, NULL
884},
885{ "title", 0, 0, 0, 0, 0, 0, 0, "document title ",
886 DECL html_pcdata, NULL, DECL i18n_attrs, NULL, NULL
887},
888{ "tr", 0, 0, 0, 0, 0, 0, 0, "table row ",
889 DECL tr_contents , "td" , DECL talign_attrs, DECL bgcolor_attr, NULL
890},
891{ "tt", 0, 3, 0, 0, 0, 0, 1, "teletype or monospaced text style",
892 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
893},
894{ "u", 0, 3, 0, 0, 1, 1, 1, "underlined text style",
895 DECL html_inline, NULL, NULL, DECL html_attrs, NULL
896},
897{ "ul", 0, 0, 0, 0, 0, 0, 0, "unordered list ",
898 DECL li_elt , "li" , DECL html_attrs, DECL ul_depr, NULL
899},
900{ "var", 0, 0, 0, 0, 0, 0, 1, "instance of a variable or program argument",
901 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
902}
903};
904
905/*
906 * start tags that imply the end of current element
907 */
908static const char * const htmlStartClose[] = {
909"form", "form", "p", "hr", "h1", "h2", "h3", "h4", "h5", "h6",
910 "dl", "ul", "ol", "menu", "dir", "address", "pre",
911 "listing", "xmp", "head", NULL,
912"head", "p", NULL,
913"title", "p", NULL,
914"body", "head", "style", "link", "title", "p", NULL,
915"frameset", "head", "style", "link", "title", "p", NULL,
916"li", "p", "h1", "h2", "h3", "h4", "h5", "h6", "dl", "address",
917 "pre", "listing", "xmp", "head", "li", NULL,
918"hr", "p", "head", NULL,
919"h1", "p", "head", NULL,
920"h2", "p", "head", NULL,
921"h3", "p", "head", NULL,
922"h4", "p", "head", NULL,
923"h5", "p", "head", NULL,
924"h6", "p", "head", NULL,
925"dir", "p", "head", NULL,
926"address", "p", "head", "ul", NULL,
927"pre", "p", "head", "ul", NULL,
928"listing", "p", "head", NULL,
929"xmp", "p", "head", NULL,
930"blockquote", "p", "head", NULL,
931"dl", "p", "dt", "menu", "dir", "address", "pre", "listing",
932 "xmp", "head", NULL,
933"dt", "p", "menu", "dir", "address", "pre", "listing", "xmp",
934 "head", "dd", NULL,
935"dd", "p", "menu", "dir", "address", "pre", "listing", "xmp",
936 "head", "dt", NULL,
937"ul", "p", "head", "ol", "menu", "dir", "address", "pre",
938 "listing", "xmp", NULL,
939"ol", "p", "head", "ul", NULL,
940"menu", "p", "head", "ul", NULL,
941"p", "p", "head", "h1", "h2", "h3", "h4", "h5", "h6", NULL,
942"div", "p", "head", NULL,
943"noscript", "p", "head", NULL,
944"center", "font", "b", "i", "p", "head", NULL,
945"a", "a", NULL,
946"caption", "p", NULL,
947"colgroup", "caption", "colgroup", "col", "p", NULL,
948"col", "caption", "col", "p", NULL,
949"table", "p", "head", "h1", "h2", "h3", "h4", "h5", "h6", "pre",
950 "listing", "xmp", "a", NULL,
951"th", "th", "td", "p", "span", "font", "a", "b", "i", "u", NULL,
952"td", "th", "td", "p", "span", "font", "a", "b", "i", "u", NULL,
953"tr", "th", "td", "tr", "caption", "col", "colgroup", "p", NULL,
954"thead", "caption", "col", "colgroup", NULL,
955"tfoot", "th", "td", "tr", "caption", "col", "colgroup", "thead",
956 "tbody", "p", NULL,
957"tbody", "th", "td", "tr", "caption", "col", "colgroup", "thead",
958 "tfoot", "tbody", "p", NULL,
959"optgroup", "option", NULL,
960"option", "option", NULL,
961"fieldset", "legend", "p", "head", "h1", "h2", "h3", "h4", "h5", "h6",
962 "pre", "listing", "xmp", "a", NULL,
963NULL
964};
965
966/*
967 * The list of HTML elements which are supposed not to have
968 * CDATA content and where a p element will be implied
969 *
970 * TODO: extend that list by reading the HTML SGML DTD on
971 * implied paragraph
972 */
973static const char *const htmlNoContentElements[] = {
974 "html",
975 "head",
976 NULL
977};
978
979/*
980 * The list of HTML attributes which are of content %Script;
981 * NOTE: when adding ones, check htmlIsScriptAttribute() since
982 * it assumes the name starts with 'on'
983 */
984static const char *const htmlScriptAttributes[] = {
985 "onclick",
986 "ondblclick",
987 "onmousedown",
988 "onmouseup",
989 "onmouseover",
990 "onmousemove",
991 "onmouseout",
992 "onkeypress",
993 "onkeydown",
994 "onkeyup",
995 "onload",
996 "onunload",
997 "onfocus",
998 "onblur",
999 "onsubmit",
1000 "onrest",
1001 "onchange",
1002 "onselect"
1003};
1004
1005/*
1006 * This table is used by the htmlparser to know what to do with
1007 * broken html pages. By assigning different priorities to different
1008 * elements the parser can decide how to handle extra endtags.
1009 * Endtags are only allowed to close elements with lower or equal
1010 * priority.
1011 */
1012
1013typedef struct {
1014 const char *name;
1015 int priority;
1016} elementPriority;
1017
1018static const elementPriority htmlEndPriority[] = {
1019 {"div", 150},
1020 {"td", 160},
1021 {"th", 160},
1022 {"tr", 170},
1023 {"thead", 180},
1024 {"tbody", 180},
1025 {"tfoot", 180},
1026 {"table", 190},
1027 {"head", 200},
1028 {"body", 200},
1029 {"html", 220},
1030 {NULL, 100} /* Default priority */
1031};
1032
1033static const char** htmlStartCloseIndex[100];
1034static int htmlStartCloseIndexinitialized = 0;
1035
1036/************************************************************************
1037 * *
1038 * functions to handle HTML specific data *
1039 * *
1040 ************************************************************************/
1041
1042/**
1043 * htmlInitAutoClose:
1044 *
1045 * Initialize the htmlStartCloseIndex for fast lookup of closing tags names.
1046 * This is not reentrant. Call xmlInitParser() once before processing in
1047 * case of use in multithreaded programs.
1048 */
1049void
1050htmlInitAutoClose(void) {
1051 int indx, i = 0;
1052
1053 if (htmlStartCloseIndexinitialized) return;
1054
1055 for (indx = 0;indx < 100;indx ++) htmlStartCloseIndex[indx] = NULL;
1056 indx = 0;
1057 while ((htmlStartClose[i] != NULL) && (indx < 100 - 1)) {
1058 htmlStartCloseIndex[indx++] = (const char**) &htmlStartClose[i];
1059 while (htmlStartClose[i] != NULL) i++;
1060 i++;
1061 }
1062 htmlStartCloseIndexinitialized = 1;
1063}
1064
1065/**
1066 * htmlTagLookup:
1067 * @tag: The tag name in lowercase
1068 *
1069 * Lookup the HTML tag in the ElementTable
1070 *
1071 * Returns the related htmlElemDescPtr or NULL if not found.
1072 */
1073const htmlElemDesc *
1074htmlTagLookup(const xmlChar *tag) {
1075 unsigned int i;
1076
1077 for (i = 0; i < (sizeof(html40ElementTable) /
1078 sizeof(html40ElementTable[0]));i++) {
1079 if (!xmlStrcasecmp(tag, BAD_CAST html40ElementTable[i].name))
1080 return((htmlElemDescPtr) &html40ElementTable[i]);
1081 }
1082 return(NULL);
1083}
1084
1085/**
1086 * htmlGetEndPriority:
1087 * @name: The name of the element to look up the priority for.
1088 *
1089 * Return value: The "endtag" priority.
1090 **/
1091static int
1092htmlGetEndPriority (const xmlChar *name) {
1093 int i = 0;
1094
1095 while ((htmlEndPriority[i].name != NULL) &&
1096 (!xmlStrEqual((const xmlChar *)htmlEndPriority[i].name, name)))
1097 i++;
1098
1099 return(htmlEndPriority[i].priority);
1100}
1101
1102
1103/**
1104 * htmlCheckAutoClose:
1105 * @newtag: The new tag name
1106 * @oldtag: The old tag name
1107 *
1108 * Checks whether the new tag is one of the registered valid tags for
1109 * closing old.
1110 * Initialize the htmlStartCloseIndex for fast lookup of closing tags names.
1111 *
1112 * Returns 0 if no, 1 if yes.
1113 */
1114static int
1115htmlCheckAutoClose(const xmlChar * newtag, const xmlChar * oldtag)
1116{
1117 int i, indx;
1118 const char **closed = NULL;
1119
1120 if (htmlStartCloseIndexinitialized == 0)
1121 htmlInitAutoClose();
1122
1123 /* inefficient, but not a big deal */
1124 for (indx = 0; indx < 100; indx++) {
1125 closed = htmlStartCloseIndex[indx];
1126 if (closed == NULL)
1127 return (0);
1128 if (xmlStrEqual(BAD_CAST * closed, newtag))
1129 break;
1130 }
1131
1132 i = closed - htmlStartClose;
1133 i++;
1134 while (htmlStartClose[i] != NULL) {
1135 if (xmlStrEqual(BAD_CAST htmlStartClose[i], oldtag)) {
1136 return (1);
1137 }
1138 i++;
1139 }
1140 return (0);
1141}
1142
1143/**
1144 * htmlAutoCloseOnClose:
1145 * @ctxt: an HTML parser context
1146 * @newtag: The new tag name
1147 * @force: force the tag closure
1148 *
1149 * The HTML DTD allows an ending tag to implicitly close other tags.
1150 */
1151static void
1152htmlAutoCloseOnClose(htmlParserCtxtPtr ctxt, const xmlChar * newtag)
1153{
1154 const htmlElemDesc *info;
1155 int i, priority;
1156
1157 priority = htmlGetEndPriority(newtag);
1158
1159 for (i = (ctxt->nameNr - 1); i >= 0; i--) {
1160
1161 if (xmlStrEqual(newtag, ctxt->nameTab[i]))
1162 break;
1163 /*
1164 * A missplaced endtag can only close elements with lower
1165 * or equal priority, so if we find an element with higher
1166 * priority before we find an element with
1167 * matching name, we just ignore this endtag
1168 */
1169 if (htmlGetEndPriority(ctxt->nameTab[i]) > priority)
1170 return;
1171 }
1172 if (i < 0)
1173 return;
1174
1175 while (!xmlStrEqual(newtag, ctxt->name)) {
1176 info = htmlTagLookup(ctxt->name);
1177 if ((info != NULL) && (info->endTag == 3)) {
1178 htmlParseErr(ctxt, XML_ERR_TAG_NAME_MISMATCH,
1179 "Opening and ending tag mismatch: %s and %s\n",
1180 newtag, ctxt->name);
1181 }
1182 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
1183 ctxt->sax->endElement(ctxt->userData, ctxt->name);
1184 htmlnamePop(ctxt);
1185 }
1186}
1187
1188/**
1189 * htmlAutoCloseOnEnd:
1190 * @ctxt: an HTML parser context
1191 *
1192 * Close all remaining tags at the end of the stream
1193 */
1194static void
1195htmlAutoCloseOnEnd(htmlParserCtxtPtr ctxt)
1196{
1197 int i;
1198
1199 if (ctxt->nameNr == 0)
1200 return;
1201 for (i = (ctxt->nameNr - 1); i >= 0; i--) {
1202 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
1203 ctxt->sax->endElement(ctxt->userData, ctxt->name);
1204 htmlnamePop(ctxt);
1205 }
1206}
1207
1208/**
1209 * htmlAutoClose:
1210 * @ctxt: an HTML parser context
1211 * @newtag: The new tag name or NULL
1212 *
1213 * The HTML DTD allows a tag to implicitly close other tags.
1214 * The list is kept in htmlStartClose array. This function is
1215 * called when a new tag has been detected and generates the
1216 * appropriates closes if possible/needed.
1217 * If newtag is NULL this mean we are at the end of the resource
1218 * and we should check
1219 */
1220static void
1221htmlAutoClose(htmlParserCtxtPtr ctxt, const xmlChar * newtag)
1222{
1223 while ((newtag != NULL) && (ctxt->name != NULL) &&
1224 (htmlCheckAutoClose(newtag, ctxt->name))) {
1225 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
1226 ctxt->sax->endElement(ctxt->userData, ctxt->name);
1227 htmlnamePop(ctxt);
1228 }
1229 if (newtag == NULL) {
1230 htmlAutoCloseOnEnd(ctxt);
1231 return;
1232 }
1233 while ((newtag == NULL) && (ctxt->name != NULL) &&
1234 ((xmlStrEqual(ctxt->name, BAD_CAST "head")) ||
1235 (xmlStrEqual(ctxt->name, BAD_CAST "body")) ||
1236 (xmlStrEqual(ctxt->name, BAD_CAST "html")))) {
1237 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
1238 ctxt->sax->endElement(ctxt->userData, ctxt->name);
1239 htmlnamePop(ctxt);
1240 }
1241}
1242
1243/**
1244 * htmlAutoCloseTag:
1245 * @doc: the HTML document
1246 * @name: The tag name
1247 * @elem: the HTML element
1248 *
1249 * The HTML DTD allows a tag to implicitly close other tags.
1250 * The list is kept in htmlStartClose array. This function checks
1251 * if the element or one of it's children would autoclose the
1252 * given tag.
1253 *
1254 * Returns 1 if autoclose, 0 otherwise
1255 */
1256int
1257htmlAutoCloseTag(htmlDocPtr doc, const xmlChar *name, htmlNodePtr elem) {
1258 htmlNodePtr child;
1259
1260 if (elem == NULL) return(1);
1261 if (xmlStrEqual(name, elem->name)) return(0);
1262 if (htmlCheckAutoClose(elem->name, name)) return(1);
1263 child = elem->children;
1264 while (child != NULL) {
1265 if (htmlAutoCloseTag(doc, name, child)) return(1);
1266 child = child->next;
1267 }
1268 return(0);
1269}
1270
1271/**
1272 * htmlIsAutoClosed:
1273 * @doc: the HTML document
1274 * @elem: the HTML element
1275 *
1276 * The HTML DTD allows a tag to implicitly close other tags.
1277 * The list is kept in htmlStartClose array. This function checks
1278 * if a tag is autoclosed by one of it's child
1279 *
1280 * Returns 1 if autoclosed, 0 otherwise
1281 */
1282int
1283htmlIsAutoClosed(htmlDocPtr doc, htmlNodePtr elem) {
1284 htmlNodePtr child;
1285
1286 if (elem == NULL) return(1);
1287 child = elem->children;
1288 while (child != NULL) {
1289 if (htmlAutoCloseTag(doc, elem->name, child)) return(1);
1290 child = child->next;
1291 }
1292 return(0);
1293}
1294
1295/**
1296 * htmlCheckImplied:
1297 * @ctxt: an HTML parser context
1298 * @newtag: The new tag name
1299 *
1300 * The HTML DTD allows a tag to exists only implicitly
1301 * called when a new tag has been detected and generates the
1302 * appropriates implicit tags if missing
1303 */
1304static void
1305htmlCheckImplied(htmlParserCtxtPtr ctxt, const xmlChar *newtag) {
1306 if (!htmlOmittedDefaultValue)
1307 return;
1308 if (xmlStrEqual(newtag, BAD_CAST"html"))
1309 return;
1310 if (ctxt->nameNr <= 0) {
1311 htmlnamePush(ctxt, BAD_CAST"html");
1312 if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
1313 ctxt->sax->startElement(ctxt->userData, BAD_CAST"html", NULL);
1314 }
1315 if ((xmlStrEqual(newtag, BAD_CAST"body")) || (xmlStrEqual(newtag, BAD_CAST"head")))
1316 return;
1317 if ((ctxt->nameNr <= 1) &&
1318 ((xmlStrEqual(newtag, BAD_CAST"script")) ||
1319 (xmlStrEqual(newtag, BAD_CAST"style")) ||
1320 (xmlStrEqual(newtag, BAD_CAST"meta")) ||
1321 (xmlStrEqual(newtag, BAD_CAST"link")) ||
1322 (xmlStrEqual(newtag, BAD_CAST"title")) ||
1323 (xmlStrEqual(newtag, BAD_CAST"base")))) {
1324 /*
1325 * dropped OBJECT ... i you put it first BODY will be
1326 * assumed !
1327 */
1328 htmlnamePush(ctxt, BAD_CAST"head");
1329 if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
1330 ctxt->sax->startElement(ctxt->userData, BAD_CAST"head", NULL);
1331 } else if ((!xmlStrEqual(newtag, BAD_CAST"noframes")) &&
1332 (!xmlStrEqual(newtag, BAD_CAST"frame")) &&
1333 (!xmlStrEqual(newtag, BAD_CAST"frameset"))) {
1334 int i;
1335 for (i = 0;i < ctxt->nameNr;i++) {
1336 if (xmlStrEqual(ctxt->nameTab[i], BAD_CAST"body")) {
1337 return;
1338 }
1339 if (xmlStrEqual(ctxt->nameTab[i], BAD_CAST"head")) {
1340 return;
1341 }
1342 }
1343
1344 htmlnamePush(ctxt, BAD_CAST"body");
1345 if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
1346 ctxt->sax->startElement(ctxt->userData, BAD_CAST"body", NULL);
1347 }
1348}
1349
1350/**
1351 * htmlCheckParagraph
1352 * @ctxt: an HTML parser context
1353 *
1354 * Check whether a p element need to be implied before inserting
1355 * characters in the current element.
1356 *
1357 * Returns 1 if a paragraph has been inserted, 0 if not and -1
1358 * in case of error.
1359 */
1360
1361static int
1362htmlCheckParagraph(htmlParserCtxtPtr ctxt) {
1363 const xmlChar *tag;
1364 int i;
1365
1366 if (ctxt == NULL)
1367 return(-1);
1368 tag = ctxt->name;
1369 if (tag == NULL) {
1370 htmlAutoClose(ctxt, BAD_CAST"p");
1371 htmlCheckImplied(ctxt, BAD_CAST"p");
1372 htmlnamePush(ctxt, BAD_CAST"p");
1373 if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
1374 ctxt->sax->startElement(ctxt->userData, BAD_CAST"p", NULL);
1375 return(1);
1376 }
1377 if (!htmlOmittedDefaultValue)
1378 return(0);
1379 for (i = 0; htmlNoContentElements[i] != NULL; i++) {
1380 if (xmlStrEqual(tag, BAD_CAST htmlNoContentElements[i])) {
1381 htmlAutoClose(ctxt, BAD_CAST"p");
1382 htmlCheckImplied(ctxt, BAD_CAST"p");
1383 htmlnamePush(ctxt, BAD_CAST"p");
1384 if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
1385 ctxt->sax->startElement(ctxt->userData, BAD_CAST"p", NULL);
1386 return(1);
1387 }
1388 }
1389 return(0);
1390}
1391
1392/**
1393 * htmlIsScriptAttribute:
1394 * @name: an attribute name
1395 *
1396 * Check if an attribute is of content type Script
1397 *
1398 * Returns 1 is the attribute is a script 0 otherwise
1399 */
1400int
1401htmlIsScriptAttribute(const xmlChar *name) {
1402 unsigned int i;
1403
1404 if (name == NULL)
1405 return(0);
1406 /*
1407 * all script attributes start with 'on'
1408 */
1409 if ((name[0] != 'o') || (name[1] != 'n'))
1410 return(0);
1411 for (i = 0;
1412 i < sizeof(htmlScriptAttributes)/sizeof(htmlScriptAttributes[0]);
1413 i++) {
1414 if (xmlStrEqual(name, (const xmlChar *) htmlScriptAttributes[i]))
1415 return(1);
1416 }
1417 return(0);
1418}
1419
1420/************************************************************************
1421 * *
1422 * The list of HTML predefined entities *
1423 * *
1424 ************************************************************************/
1425
1426
1427static const htmlEntityDesc html40EntitiesTable[] = {
1428/*
1429 * the 4 absolute ones, plus apostrophe.
1430 */
1431{ 34, "quot", "quotation mark = APL quote, U+0022 ISOnum" },
1432{ 38, "amp", "ampersand, U+0026 ISOnum" },
1433{ 39, "apos", "single quote" },
1434{ 60, "lt", "less-than sign, U+003C ISOnum" },
1435{ 62, "gt", "greater-than sign, U+003E ISOnum" },
1436
1437/*
1438 * A bunch still in the 128-255 range
1439 * Replacing them depend really on the charset used.
1440 */
1441{ 160, "nbsp", "no-break space = non-breaking space, U+00A0 ISOnum" },
1442{ 161, "iexcl","inverted exclamation mark, U+00A1 ISOnum" },
1443{ 162, "cent", "cent sign, U+00A2 ISOnum" },
1444{ 163, "pound","pound sign, U+00A3 ISOnum" },
1445{ 164, "curren","currency sign, U+00A4 ISOnum" },
1446{ 165, "yen", "yen sign = yuan sign, U+00A5 ISOnum" },
1447{ 166, "brvbar","broken bar = broken vertical bar, U+00A6 ISOnum" },
1448{ 167, "sect", "section sign, U+00A7 ISOnum" },
1449{ 168, "uml", "diaeresis = spacing diaeresis, U+00A8 ISOdia" },
1450{ 169, "copy", "copyright sign, U+00A9 ISOnum" },
1451{ 170, "ordf", "feminine ordinal indicator, U+00AA ISOnum" },
1452{ 171, "laquo","left-pointing double angle quotation mark = left pointing guillemet, U+00AB ISOnum" },
1453{ 172, "not", "not sign, U+00AC ISOnum" },
1454{ 173, "shy", "soft hyphen = discretionary hyphen, U+00AD ISOnum" },
1455{ 174, "reg", "registered sign = registered trade mark sign, U+00AE ISOnum" },
1456{ 175, "macr", "macron = spacing macron = overline = APL overbar, U+00AF ISOdia" },
1457{ 176, "deg", "degree sign, U+00B0 ISOnum" },
1458{ 177, "plusmn","plus-minus sign = plus-or-minus sign, U+00B1 ISOnum" },
1459{ 178, "sup2", "superscript two = superscript digit two = squared, U+00B2 ISOnum" },
1460{ 179, "sup3", "superscript three = superscript digit three = cubed, U+00B3 ISOnum" },
1461{ 180, "acute","acute accent = spacing acute, U+00B4 ISOdia" },
1462{ 181, "micro","micro sign, U+00B5 ISOnum" },
1463{ 182, "para", "pilcrow sign = paragraph sign, U+00B6 ISOnum" },
1464{ 183, "middot","middle dot = Georgian comma Greek middle dot, U+00B7 ISOnum" },
1465{ 184, "cedil","cedilla = spacing cedilla, U+00B8 ISOdia" },
1466{ 185, "sup1", "superscript one = superscript digit one, U+00B9 ISOnum" },
1467{ 186, "ordm", "masculine ordinal indicator, U+00BA ISOnum" },
1468{ 187, "raquo","right-pointing double angle quotation mark right pointing guillemet, U+00BB ISOnum" },
1469{ 188, "frac14","vulgar fraction one quarter = fraction one quarter, U+00BC ISOnum" },
1470{ 189, "frac12","vulgar fraction one half = fraction one half, U+00BD ISOnum" },
1471{ 190, "frac34","vulgar fraction three quarters = fraction three quarters, U+00BE ISOnum" },
1472{ 191, "iquest","inverted question mark = turned question mark, U+00BF ISOnum" },
1473{ 192, "Agrave","latin capital letter A with grave = latin capital letter A grave, U+00C0 ISOlat1" },
1474{ 193, "Aacute","latin capital letter A with acute, U+00C1 ISOlat1" },
1475{ 194, "Acirc","latin capital letter A with circumflex, U+00C2 ISOlat1" },
1476{ 195, "Atilde","latin capital letter A with tilde, U+00C3 ISOlat1" },
1477{ 196, "Auml", "latin capital letter A with diaeresis, U+00C4 ISOlat1" },
1478{ 197, "Aring","latin capital letter A with ring above = latin capital letter A ring, U+00C5 ISOlat1" },
1479{ 198, "AElig","latin capital letter AE = latin capital ligature AE, U+00C6 ISOlat1" },
1480{ 199, "Ccedil","latin capital letter C with cedilla, U+00C7 ISOlat1" },
1481{ 200, "Egrave","latin capital letter E with grave, U+00C8 ISOlat1" },
1482{ 201, "Eacute","latin capital letter E with acute, U+00C9 ISOlat1" },
1483{ 202, "Ecirc","latin capital letter E with circumflex, U+00CA ISOlat1" },
1484{ 203, "Euml", "latin capital letter E with diaeresis, U+00CB ISOlat1" },
1485{ 204, "Igrave","latin capital letter I with grave, U+00CC ISOlat1" },
1486{ 205, "Iacute","latin capital letter I with acute, U+00CD ISOlat1" },
1487{ 206, "Icirc","latin capital letter I with circumflex, U+00CE ISOlat1" },
1488{ 207, "Iuml", "latin capital letter I with diaeresis, U+00CF ISOlat1" },
1489{ 208, "ETH", "latin capital letter ETH, U+00D0 ISOlat1" },
1490{ 209, "Ntilde","latin capital letter N with tilde, U+00D1 ISOlat1" },
1491{ 210, "Ograve","latin capital letter O with grave, U+00D2 ISOlat1" },
1492{ 211, "Oacute","latin capital letter O with acute, U+00D3 ISOlat1" },
1493{ 212, "Ocirc","latin capital letter O with circumflex, U+00D4 ISOlat1" },
1494{ 213, "Otilde","latin capital letter O with tilde, U+00D5 ISOlat1" },
1495{ 214, "Ouml", "latin capital letter O with diaeresis, U+00D6 ISOlat1" },
1496{ 215, "times","multiplication sign, U+00D7 ISOnum" },
1497{ 216, "Oslash","latin capital letter O with stroke latin capital letter O slash, U+00D8 ISOlat1" },
1498{ 217, "Ugrave","latin capital letter U with grave, U+00D9 ISOlat1" },
1499{ 218, "Uacute","latin capital letter U with acute, U+00DA ISOlat1" },
1500{ 219, "Ucirc","latin capital letter U with circumflex, U+00DB ISOlat1" },
1501{ 220, "Uuml", "latin capital letter U with diaeresis, U+00DC ISOlat1" },
1502{ 221, "Yacute","latin capital letter Y with acute, U+00DD ISOlat1" },
1503{ 222, "THORN","latin capital letter THORN, U+00DE ISOlat1" },
1504{ 223, "szlig","latin small letter sharp s = ess-zed, U+00DF ISOlat1" },
1505{ 224, "agrave","latin small letter a with grave = latin small letter a grave, U+00E0 ISOlat1" },
1506{ 225, "aacute","latin small letter a with acute, U+00E1 ISOlat1" },
1507{ 226, "acirc","latin small letter a with circumflex, U+00E2 ISOlat1" },
1508{ 227, "atilde","latin small letter a with tilde, U+00E3 ISOlat1" },
1509{ 228, "auml", "latin small letter a with diaeresis, U+00E4 ISOlat1" },
1510{ 229, "aring","latin small letter a with ring above = latin small letter a ring, U+00E5 ISOlat1" },
1511{ 230, "aelig","latin small letter ae = latin small ligature ae, U+00E6 ISOlat1" },
1512{ 231, "ccedil","latin small letter c with cedilla, U+00E7 ISOlat1" },
1513{ 232, "egrave","latin small letter e with grave, U+00E8 ISOlat1" },
1514{ 233, "eacute","latin small letter e with acute, U+00E9 ISOlat1" },
1515{ 234, "ecirc","latin small letter e with circumflex, U+00EA ISOlat1" },
1516{ 235, "euml", "latin small letter e with diaeresis, U+00EB ISOlat1" },
1517{ 236, "igrave","latin small letter i with grave, U+00EC ISOlat1" },
1518{ 237, "iacute","latin small letter i with acute, U+00ED ISOlat1" },
1519{ 238, "icirc","latin small letter i with circumflex, U+00EE ISOlat1" },
1520{ 239, "iuml", "latin small letter i with diaeresis, U+00EF ISOlat1" },
1521{ 240, "eth", "latin small letter eth, U+00F0 ISOlat1" },
1522{ 241, "ntilde","latin small letter n with tilde, U+00F1 ISOlat1" },
1523{ 242, "ograve","latin small letter o with grave, U+00F2 ISOlat1" },
1524{ 243, "oacute","latin small letter o with acute, U+00F3 ISOlat1" },
1525{ 244, "ocirc","latin small letter o with circumflex, U+00F4 ISOlat1" },
1526{ 245, "otilde","latin small letter o with tilde, U+00F5 ISOlat1" },
1527{ 246, "ouml", "latin small letter o with diaeresis, U+00F6 ISOlat1" },
1528{ 247, "divide","division sign, U+00F7 ISOnum" },
1529{ 248, "oslash","latin small letter o with stroke, = latin small letter o slash, U+00F8 ISOlat1" },
1530{ 249, "ugrave","latin small letter u with grave, U+00F9 ISOlat1" },
1531{ 250, "uacute","latin small letter u with acute, U+00FA ISOlat1" },
1532{ 251, "ucirc","latin small letter u with circumflex, U+00FB ISOlat1" },
1533{ 252, "uuml", "latin small letter u with diaeresis, U+00FC ISOlat1" },
1534{ 253, "yacute","latin small letter y with acute, U+00FD ISOlat1" },
1535{ 254, "thorn","latin small letter thorn with, U+00FE ISOlat1" },
1536{ 255, "yuml", "latin small letter y with diaeresis, U+00FF ISOlat1" },
1537
1538{ 338, "OElig","latin capital ligature OE, U+0152 ISOlat2" },
1539{ 339, "oelig","latin small ligature oe, U+0153 ISOlat2" },
1540{ 352, "Scaron","latin capital letter S with caron, U+0160 ISOlat2" },
1541{ 353, "scaron","latin small letter s with caron, U+0161 ISOlat2" },
1542{ 376, "Yuml", "latin capital letter Y with diaeresis, U+0178 ISOlat2" },
1543
1544/*
1545 * Anything below should really be kept as entities references
1546 */
1547{ 402, "fnof", "latin small f with hook = function = florin, U+0192 ISOtech" },
1548
1549{ 710, "circ", "modifier letter circumflex accent, U+02C6 ISOpub" },
1550{ 732, "tilde","small tilde, U+02DC ISOdia" },
1551
1552{ 913, "Alpha","greek capital letter alpha, U+0391" },
1553{ 914, "Beta", "greek capital letter beta, U+0392" },
1554{ 915, "Gamma","greek capital letter gamma, U+0393 ISOgrk3" },
1555{ 916, "Delta","greek capital letter delta, U+0394 ISOgrk3" },
1556{ 917, "Epsilon","greek capital letter epsilon, U+0395" },
1557{ 918, "Zeta", "greek capital letter zeta, U+0396" },
1558{ 919, "Eta", "greek capital letter eta, U+0397" },
1559{ 920, "Theta","greek capital letter theta, U+0398 ISOgrk3" },
1560{ 921, "Iota", "greek capital letter iota, U+0399" },
1561{ 922, "Kappa","greek capital letter kappa, U+039A" },
1562{ 923, "Lambda", "greek capital letter lambda, U+039B ISOgrk3" },
1563{ 924, "Mu", "greek capital letter mu, U+039C" },
1564{ 925, "Nu", "greek capital letter nu, U+039D" },
1565{ 926, "Xi", "greek capital letter xi, U+039E ISOgrk3" },
1566{ 927, "Omicron","greek capital letter omicron, U+039F" },
1567{ 928, "Pi", "greek capital letter pi, U+03A0 ISOgrk3" },
1568{ 929, "Rho", "greek capital letter rho, U+03A1" },
1569{ 931, "Sigma","greek capital letter sigma, U+03A3 ISOgrk3" },
1570{ 932, "Tau", "greek capital letter tau, U+03A4" },
1571{ 933, "Upsilon","greek capital letter upsilon, U+03A5 ISOgrk3" },
1572{ 934, "Phi", "greek capital letter phi, U+03A6 ISOgrk3" },
1573{ 935, "Chi", "greek capital letter chi, U+03A7" },
1574{ 936, "Psi", "greek capital letter psi, U+03A8 ISOgrk3" },
1575{ 937, "Omega","greek capital letter omega, U+03A9 ISOgrk3" },
1576
1577{ 945, "alpha","greek small letter alpha, U+03B1 ISOgrk3" },
1578{ 946, "beta", "greek small letter beta, U+03B2 ISOgrk3" },
1579{ 947, "gamma","greek small letter gamma, U+03B3 ISOgrk3" },
1580{ 948, "delta","greek small letter delta, U+03B4 ISOgrk3" },
1581{ 949, "epsilon","greek small letter epsilon, U+03B5 ISOgrk3" },
1582{ 950, "zeta", "greek small letter zeta, U+03B6 ISOgrk3" },
1583{ 951, "eta", "greek small letter eta, U+03B7 ISOgrk3" },
1584{ 952, "theta","greek small letter theta, U+03B8 ISOgrk3" },
1585{ 953, "iota", "greek small letter iota, U+03B9 ISOgrk3" },
1586{ 954, "kappa","greek small letter kappa, U+03BA ISOgrk3" },
1587{ 955, "lambda","greek small letter lambda, U+03BB ISOgrk3" },
1588{ 956, "mu", "greek small letter mu, U+03BC ISOgrk3" },
1589{ 957, "nu", "greek small letter nu, U+03BD ISOgrk3" },
1590{ 958, "xi", "greek small letter xi, U+03BE ISOgrk3" },
1591{ 959, "omicron","greek small letter omicron, U+03BF NEW" },
1592{ 960, "pi", "greek small letter pi, U+03C0 ISOgrk3" },
1593{ 961, "rho", "greek small letter rho, U+03C1 ISOgrk3" },
1594{ 962, "sigmaf","greek small letter final sigma, U+03C2 ISOgrk3" },
1595{ 963, "sigma","greek small letter sigma, U+03C3 ISOgrk3" },
1596{ 964, "tau", "greek small letter tau, U+03C4 ISOgrk3" },
1597{ 965, "upsilon","greek small letter upsilon, U+03C5 ISOgrk3" },
1598{ 966, "phi", "greek small letter phi, U+03C6 ISOgrk3" },
1599{ 967, "chi", "greek small letter chi, U+03C7 ISOgrk3" },
1600{ 968, "psi", "greek small letter psi, U+03C8 ISOgrk3" },
1601{ 969, "omega","greek small letter omega, U+03C9 ISOgrk3" },
1602{ 977, "thetasym","greek small letter theta symbol, U+03D1 NEW" },
1603{ 978, "upsih","greek upsilon with hook symbol, U+03D2 NEW" },
1604{ 982, "piv", "greek pi symbol, U+03D6 ISOgrk3" },
1605
1606{ 8194, "ensp", "en space, U+2002 ISOpub" },
1607{ 8195, "emsp", "em space, U+2003 ISOpub" },
1608{ 8201, "thinsp","thin space, U+2009 ISOpub" },
1609{ 8204, "zwnj", "zero width non-joiner, U+200C NEW RFC 2070" },
1610{ 8205, "zwj", "zero width joiner, U+200D NEW RFC 2070" },
1611{ 8206, "lrm", "left-to-right mark, U+200E NEW RFC 2070" },
1612{ 8207, "rlm", "right-to-left mark, U+200F NEW RFC 2070" },
1613{ 8211, "ndash","en dash, U+2013 ISOpub" },
1614{ 8212, "mdash","em dash, U+2014 ISOpub" },
1615{ 8216, "lsquo","left single quotation mark, U+2018 ISOnum" },
1616{ 8217, "rsquo","right single quotation mark, U+2019 ISOnum" },
1617{ 8218, "sbquo","single low-9 quotation mark, U+201A NEW" },
1618{ 8220, "ldquo","left double quotation mark, U+201C ISOnum" },
1619{ 8221, "rdquo","right double quotation mark, U+201D ISOnum" },
1620{ 8222, "bdquo","double low-9 quotation mark, U+201E NEW" },
1621{ 8224, "dagger","dagger, U+2020 ISOpub" },
1622{ 8225, "Dagger","double dagger, U+2021 ISOpub" },
1623
1624{ 8226, "bull", "bullet = black small circle, U+2022 ISOpub" },
1625{ 8230, "hellip","horizontal ellipsis = three dot leader, U+2026 ISOpub" },
1626
1627{ 8240, "permil","per mille sign, U+2030 ISOtech" },
1628
1629{ 8242, "prime","prime = minutes = feet, U+2032 ISOtech" },
1630{ 8243, "Prime","double prime = seconds = inches, U+2033 ISOtech" },
1631
1632{ 8249, "lsaquo","single left-pointing angle quotation mark, U+2039 ISO proposed" },
1633{ 8250, "rsaquo","single right-pointing angle quotation mark, U+203A ISO proposed" },
1634
1635{ 8254, "oline","overline = spacing overscore, U+203E NEW" },
1636{ 8260, "frasl","fraction slash, U+2044 NEW" },
1637
1638{ 8364, "euro", "euro sign, U+20AC NEW" },
1639
1640{ 8465, "image","blackletter capital I = imaginary part, U+2111 ISOamso" },
1641{ 8472, "weierp","script capital P = power set = Weierstrass p, U+2118 ISOamso" },
1642{ 8476, "real", "blackletter capital R = real part symbol, U+211C ISOamso" },
1643{ 8482, "trade","trade mark sign, U+2122 ISOnum" },
1644{ 8501, "alefsym","alef symbol = first transfinite cardinal, U+2135 NEW" },
1645{ 8592, "larr", "leftwards arrow, U+2190 ISOnum" },
1646{ 8593, "uarr", "upwards arrow, U+2191 ISOnum" },
1647{ 8594, "rarr", "rightwards arrow, U+2192 ISOnum" },
1648{ 8595, "darr", "downwards arrow, U+2193 ISOnum" },
1649{ 8596, "harr", "left right arrow, U+2194 ISOamsa" },
1650{ 8629, "crarr","downwards arrow with corner leftwards = carriage return, U+21B5 NEW" },
1651{ 8656, "lArr", "leftwards double arrow, U+21D0 ISOtech" },
1652{ 8657, "uArr", "upwards double arrow, U+21D1 ISOamsa" },
1653{ 8658, "rArr", "rightwards double arrow, U+21D2 ISOtech" },
1654{ 8659, "dArr", "downwards double arrow, U+21D3 ISOamsa" },
1655{ 8660, "hArr", "left right double arrow, U+21D4 ISOamsa" },
1656
1657{ 8704, "forall","for all, U+2200 ISOtech" },
1658{ 8706, "part", "partial differential, U+2202 ISOtech" },
1659{ 8707, "exist","there exists, U+2203 ISOtech" },
1660{ 8709, "empty","empty set = null set = diameter, U+2205 ISOamso" },
1661{ 8711, "nabla","nabla = backward difference, U+2207 ISOtech" },
1662{ 8712, "isin", "element of, U+2208 ISOtech" },
1663{ 8713, "notin","not an element of, U+2209 ISOtech" },
1664{ 8715, "ni", "contains as member, U+220B ISOtech" },
1665{ 8719, "prod", "n-ary product = product sign, U+220F ISOamsb" },
1666{ 8721, "sum", "n-ary summation, U+2211 ISOamsb" },
1667{ 8722, "minus","minus sign, U+2212 ISOtech" },
1668{ 8727, "lowast","asterisk operator, U+2217 ISOtech" },
1669{ 8730, "radic","square root = radical sign, U+221A ISOtech" },
1670{ 8733, "prop", "proportional to, U+221D ISOtech" },
1671{ 8734, "infin","infinity, U+221E ISOtech" },
1672{ 8736, "ang", "angle, U+2220 ISOamso" },
1673{ 8743, "and", "logical and = wedge, U+2227 ISOtech" },
1674{ 8744, "or", "logical or = vee, U+2228 ISOtech" },
1675{ 8745, "cap", "intersection = cap, U+2229 ISOtech" },
1676{ 8746, "cup", "union = cup, U+222A ISOtech" },
1677{ 8747, "int", "integral, U+222B ISOtech" },
1678{ 8756, "there4","therefore, U+2234 ISOtech" },
1679{ 8764, "sim", "tilde operator = varies with = similar to, U+223C ISOtech" },
1680{ 8773, "cong", "approximately equal to, U+2245 ISOtech" },
1681{ 8776, "asymp","almost equal to = asymptotic to, U+2248 ISOamsr" },
1682{ 8800, "ne", "not equal to, U+2260 ISOtech" },
1683{ 8801, "equiv","identical to, U+2261 ISOtech" },
1684{ 8804, "le", "less-than or equal to, U+2264 ISOtech" },
1685{ 8805, "ge", "greater-than or equal to, U+2265 ISOtech" },
1686{ 8834, "sub", "subset of, U+2282 ISOtech" },
1687{ 8835, "sup", "superset of, U+2283 ISOtech" },
1688{ 8836, "nsub", "not a subset of, U+2284 ISOamsn" },
1689{ 8838, "sube", "subset of or equal to, U+2286 ISOtech" },
1690{ 8839, "supe", "superset of or equal to, U+2287 ISOtech" },
1691{ 8853, "oplus","circled plus = direct sum, U+2295 ISOamsb" },
1692{ 8855, "otimes","circled times = vector product, U+2297 ISOamsb" },
1693{ 8869, "perp", "up tack = orthogonal to = perpendicular, U+22A5 ISOtech" },
1694{ 8901, "sdot", "dot operator, U+22C5 ISOamsb" },
1695{ 8968, "lceil","left ceiling = apl upstile, U+2308 ISOamsc" },
1696{ 8969, "rceil","right ceiling, U+2309 ISOamsc" },
1697{ 8970, "lfloor","left floor = apl downstile, U+230A ISOamsc" },
1698{ 8971, "rfloor","right floor, U+230B ISOamsc" },
1699{ 9001, "lang", "left-pointing angle bracket = bra, U+2329 ISOtech" },
1700{ 9002, "rang", "right-pointing angle bracket = ket, U+232A ISOtech" },
1701{ 9674, "loz", "lozenge, U+25CA ISOpub" },
1702
1703{ 9824, "spades","black spade suit, U+2660 ISOpub" },
1704{ 9827, "clubs","black club suit = shamrock, U+2663 ISOpub" },
1705{ 9829, "hearts","black heart suit = valentine, U+2665 ISOpub" },
1706{ 9830, "diams","black diamond suit, U+2666 ISOpub" },
1707
1708};
1709
1710/************************************************************************
1711 * *
1712 * Commodity functions to handle entities *
1713 * *
1714 ************************************************************************/
1715
1716/*
1717 * Macro used to grow the current buffer.
1718 */
1719#define growBuffer(buffer) { \
1720 xmlChar *tmp; \
1721 buffer##_size *= 2; \
1722 tmp = (xmlChar *) xmlRealloc(buffer, buffer##_size * sizeof(xmlChar)); \
1723 if (tmp == NULL) { \
1724 htmlErrMemory(ctxt, "growing buffer\n"); \
1725 xmlFree(buffer); \
1726 return(NULL); \
1727 } \
1728 buffer = tmp; \
1729}
1730
1731/**
1732 * htmlEntityLookup:
1733 * @name: the entity name
1734 *
1735 * Lookup the given entity in EntitiesTable
1736 *
1737 * TODO: the linear scan is really ugly, an hash table is really needed.
1738 *
1739 * Returns the associated htmlEntityDescPtr if found, NULL otherwise.
1740 */
1741const htmlEntityDesc *
1742htmlEntityLookup(const xmlChar *name) {
1743 unsigned int i;
1744
1745 for (i = 0;i < (sizeof(html40EntitiesTable)/
1746 sizeof(html40EntitiesTable[0]));i++) {
1747 if (xmlStrEqual(name, BAD_CAST html40EntitiesTable[i].name)) {
1748 return((htmlEntityDescPtr) &html40EntitiesTable[i]);
1749 }
1750 }
1751 return(NULL);
1752}
1753
1754/**
1755 * htmlEntityValueLookup:
1756 * @value: the entity's unicode value
1757 *
1758 * Lookup the given entity in EntitiesTable
1759 *
1760 * TODO: the linear scan is really ugly, an hash table is really needed.
1761 *
1762 * Returns the associated htmlEntityDescPtr if found, NULL otherwise.
1763 */
1764const htmlEntityDesc *
1765htmlEntityValueLookup(unsigned int value) {
1766 unsigned int i;
1767
1768 for (i = 0;i < (sizeof(html40EntitiesTable)/
1769 sizeof(html40EntitiesTable[0]));i++) {
1770 if (html40EntitiesTable[i].value >= value) {
1771 if (html40EntitiesTable[i].value > value)
1772 break;
1773 return((htmlEntityDescPtr) &html40EntitiesTable[i]);
1774 }
1775 }
1776 return(NULL);
1777}
1778
1779/**
1780 * UTF8ToHtml:
1781 * @out: a pointer to an array of bytes to store the result
1782 * @outlen: the length of @out
1783 * @in: a pointer to an array of UTF-8 chars
1784 * @inlen: the length of @in
1785 *
1786 * Take a block of UTF-8 chars in and try to convert it to an ASCII
1787 * plus HTML entities block of chars out.
1788 *
1789 * Returns 0 if success, -2 if the transcoding fails, or -1 otherwise
1790 * The value of @inlen after return is the number of octets consumed
1791 * as the return value is positive, else unpredictable.
1792 * The value of @outlen after return is the number of octets consumed.
1793 */
1794int
1795UTF8ToHtml(unsigned char* out, int *outlen,
1796 const unsigned char* in, int *inlen) {
1797 const unsigned char* processed = in;
1798 const unsigned char* outend;
1799 const unsigned char* outstart = out;
1800 const unsigned char* instart = in;
1801 const unsigned char* inend;
1802 unsigned int c, d;
1803 int trailing;
1804
1805 if ((out == NULL) || (outlen == NULL) || (inlen == NULL)) return(-1);
1806 if (in == NULL) {
1807 /*
1808 * initialization nothing to do
1809 */
1810 *outlen = 0;
1811 *inlen = 0;
1812 return(0);
1813 }
1814 inend = in + (*inlen);
1815 outend = out + (*outlen);
1816 while (in < inend) {
1817 d = *in++;
1818 if (d < 0x80) { c= d; trailing= 0; }
1819 else if (d < 0xC0) {
1820 /* trailing byte in leading position */
1821 *outlen = out - outstart;
1822 *inlen = processed - instart;
1823 return(-2);
1824 } else if (d < 0xE0) { c= d & 0x1F; trailing= 1; }
1825 else if (d < 0xF0) { c= d & 0x0F; trailing= 2; }
1826 else if (d < 0xF8) { c= d & 0x07; trailing= 3; }
1827 else {
1828 /* no chance for this in Ascii */
1829 *outlen = out - outstart;
1830 *inlen = processed - instart;
1831 return(-2);
1832 }
1833
1834 if (inend - in < trailing) {
1835 break;
1836 }
1837
1838 for ( ; trailing; trailing--) {
1839 if ((in >= inend) || (((d= *in++) & 0xC0) != 0x80))
1840 break;
1841 c <<= 6;
1842 c |= d & 0x3F;
1843 }
1844
1845 /* assertion: c is a single UTF-4 value */
1846 if (c < 0x80) {
1847 if (out + 1 >= outend)
1848 break;
1849 *out++ = c;
1850 } else {
1851 int len;
1852 const htmlEntityDesc * ent;
1853 const char *cp;
1854 char nbuf[16];
1855
1856 /*
1857 * Try to lookup a predefined HTML entity for it
1858 */
1859
1860 ent = htmlEntityValueLookup(c);
1861 if (ent == NULL) {
1862 snprintf(nbuf, sizeof(nbuf), "#%u", c);
1863 cp = nbuf;
1864 }
1865 else
1866 cp = ent->name;
1867 len = strlen(cp);
1868 if (out + 2 + len >= outend)
1869 break;
1870 *out++ = '&';
1871 memcpy(out, cp, len);
1872 out += len;
1873 *out++ = ';';
1874 }
1875 processed = in;
1876 }
1877 *outlen = out - outstart;
1878 *inlen = processed - instart;
1879 return(0);
1880}
1881
1882/**
1883 * htmlEncodeEntities:
1884 * @out: a pointer to an array of bytes to store the result
1885 * @outlen: the length of @out
1886 * @in: a pointer to an array of UTF-8 chars
1887 * @inlen: the length of @in
1888 * @quoteChar: the quote character to escape (' or ") or zero.
1889 *
1890 * Take a block of UTF-8 chars in and try to convert it to an ASCII
1891 * plus HTML entities block of chars out.
1892 *
1893 * Returns 0 if success, -2 if the transcoding fails, or -1 otherwise
1894 * The value of @inlen after return is the number of octets consumed
1895 * as the return value is positive, else unpredictable.
1896 * The value of @outlen after return is the number of octets consumed.
1897 */
1898int
1899htmlEncodeEntities(unsigned char* out, int *outlen,
1900 const unsigned char* in, int *inlen, int quoteChar) {
1901 const unsigned char* processed = in;
1902 const unsigned char* outend;
1903 const unsigned char* outstart = out;
1904 const unsigned char* instart = in;
1905 const unsigned char* inend;
1906 unsigned int c, d;
1907 int trailing;
1908
1909 if ((out == NULL) || (outlen == NULL) || (inlen == NULL) || (in == NULL))
1910 return(-1);
1911 outend = out + (*outlen);
1912 inend = in + (*inlen);
1913 while (in < inend) {
1914 d = *in++;
1915 if (d < 0x80) { c= d; trailing= 0; }
1916 else if (d < 0xC0) {
1917 /* trailing byte in leading position */
1918 *outlen = out - outstart;
1919 *inlen = processed - instart;
1920 return(-2);
1921 } else if (d < 0xE0) { c= d & 0x1F; trailing= 1; }
1922 else if (d < 0xF0) { c= d & 0x0F; trailing= 2; }
1923 else if (d < 0xF8) { c= d & 0x07; trailing= 3; }
1924 else {
1925 /* no chance for this in Ascii */
1926 *outlen = out - outstart;
1927 *inlen = processed - instart;
1928 return(-2);
1929 }
1930
1931 if (inend - in < trailing)
1932 break;
1933
1934 while (trailing--) {
1935 if (((d= *in++) & 0xC0) != 0x80) {
1936 *outlen = out - outstart;
1937 *inlen = processed - instart;
1938 return(-2);
1939 }
1940 c <<= 6;
1941 c |= d & 0x3F;
1942 }
1943
1944 /* assertion: c is a single UTF-4 value */
1945 if ((c < 0x80) && (c != (unsigned int) quoteChar) &&
1946 (c != '&') && (c != '<') && (c != '>')) {
1947 if (out >= outend)
1948 break;
1949 *out++ = c;
1950 } else {
1951 const htmlEntityDesc * ent;
1952 const char *cp;
1953 char nbuf[16];
1954 int len;
1955
1956 /*
1957 * Try to lookup a predefined HTML entity for it
1958 */
1959 ent = htmlEntityValueLookup(c);
1960 if (ent == NULL) {
1961 snprintf(nbuf, sizeof(nbuf), "#%u", c);
1962 cp = nbuf;
1963 }
1964 else
1965 cp = ent->name;
1966 len = strlen(cp);
1967 if (out + 2 + len > outend)
1968 break;
1969 *out++ = '&';
1970 memcpy(out, cp, len);
1971 out += len;
1972 *out++ = ';';
1973 }
1974 processed = in;
1975 }
1976 *outlen = out - outstart;
1977 *inlen = processed - instart;
1978 return(0);
1979}
1980
1981/************************************************************************
1982 * *
1983 * Commodity functions to handle streams *
1984 * *
1985 ************************************************************************/
1986
1987/**
1988 * htmlNewInputStream:
1989 * @ctxt: an HTML parser context
1990 *
1991 * Create a new input stream structure
1992 * Returns the new input stream or NULL
1993 */
1994static htmlParserInputPtr
1995htmlNewInputStream(htmlParserCtxtPtr ctxt) {
1996 htmlParserInputPtr input;
1997
1998 input = (xmlParserInputPtr) xmlMalloc(sizeof(htmlParserInput));
1999 if (input == NULL) {
2000 htmlErrMemory(ctxt, "couldn't allocate a new input stream\n");
2001 return(NULL);
2002 }
2003 memset(input, 0, sizeof(htmlParserInput));
2004 input->filename = NULL;
2005 input->directory = NULL;
2006 input->base = NULL;
2007 input->cur = NULL;
2008 input->buf = NULL;
2009 input->line = 1;
2010 input->col = 1;
2011 input->buf = NULL;
2012 input->free = NULL;
2013 input->version = NULL;
2014 input->consumed = 0;
2015 input->length = 0;
2016 return(input);
2017}
2018
2019
2020/************************************************************************
2021 * *
2022 * Commodity functions, cleanup needed ? *
2023 * *
2024 ************************************************************************/
2025/*
2026 * all tags allowing pc data from the html 4.01 loose dtd
2027 * NOTE: it might be more apropriate to integrate this information
2028 * into the html40ElementTable array but I don't want to risk any
2029 * binary incomptibility
2030 */
2031static const char *allowPCData[] = {
2032 "a", "abbr", "acronym", "address", "applet", "b", "bdo", "big",
2033 "blockquote", "body", "button", "caption", "center", "cite", "code",
2034 "dd", "del", "dfn", "div", "dt", "em", "font", "form", "h1", "h2",
2035 "h3", "h4", "h5", "h6", "i", "iframe", "ins", "kbd", "label", "legend",
2036 "li", "noframes", "noscript", "object", "p", "pre", "q", "s", "samp",
2037 "small", "span", "strike", "strong", "td", "th", "tt", "u", "var"
2038};
2039
2040/**
2041 * areBlanks:
2042 * @ctxt: an HTML parser context
2043 * @str: a xmlChar *
2044 * @len: the size of @str
2045 *
2046 * Is this a sequence of blank chars that one can ignore ?
2047 *
2048 * Returns 1 if ignorable 0 otherwise.
2049 */
2050
2051static int areBlanks(htmlParserCtxtPtr ctxt, const xmlChar *str, int len) {
2052 unsigned int i;
2053 int j;
2054 xmlNodePtr lastChild;
2055 xmlDtdPtr dtd;
2056
2057 for (j = 0;j < len;j++)
2058 if (!(IS_BLANK_CH(str[j]))) return(0);
2059
2060 if (CUR == 0) return(1);
2061 if (CUR != '<') return(0);
2062 if (ctxt->name == NULL)
2063 return(1);
2064 if (xmlStrEqual(ctxt->name, BAD_CAST"html"))
2065 return(1);
2066 if (xmlStrEqual(ctxt->name, BAD_CAST"head"))
2067 return(1);
2068
2069 /* Only strip CDATA children of the body tag for strict HTML DTDs */
2070 if (xmlStrEqual(ctxt->name, BAD_CAST "body") && ctxt->myDoc != NULL) {
2071 dtd = xmlGetIntSubset(ctxt->myDoc);
2072 if (dtd != NULL && dtd->ExternalID != NULL) {
2073 if (!xmlStrcasecmp(dtd->ExternalID, BAD_CAST "-//W3C//DTD HTML 4.01//EN") ||
2074 !xmlStrcasecmp(dtd->ExternalID, BAD_CAST "-//W3C//DTD HTML 4//EN"))
2075 return(1);
2076 }
2077 }
2078
2079 if (ctxt->node == NULL) return(0);
2080 lastChild = xmlGetLastChild(ctxt->node);
2081 while ((lastChild) && (lastChild->type == XML_COMMENT_NODE))
2082 lastChild = lastChild->prev;
2083 if (lastChild == NULL) {
2084 if ((ctxt->node->type != XML_ELEMENT_NODE) &&
2085 (ctxt->node->content != NULL)) return(0);
2086 /* keep ws in constructs like ...<b> </b>...
2087 for all tags "b" allowing PCDATA */
2088 for ( i = 0; i < sizeof(allowPCData)/sizeof(allowPCData[0]); i++ ) {
2089 if ( xmlStrEqual(ctxt->name, BAD_CAST allowPCData[i]) ) {
2090 return(0);
2091 }
2092 }
2093 } else if (xmlNodeIsText(lastChild)) {
2094 return(0);
2095 } else {
2096 /* keep ws in constructs like <p><b>xy</b> <i>z</i><p>
2097 for all tags "p" allowing PCDATA */
2098 for ( i = 0; i < sizeof(allowPCData)/sizeof(allowPCData[0]); i++ ) {
2099 if ( xmlStrEqual(lastChild->name, BAD_CAST allowPCData[i]) ) {
2100 return(0);
2101 }
2102 }
2103 }
2104 return(1);
2105}
2106
2107/**
2108 * htmlNewDocNoDtD:
2109 * @URI: URI for the dtd, or NULL
2110 * @ExternalID: the external ID of the DTD, or NULL
2111 *
2112 * Creates a new HTML document without a DTD node if @URI and @ExternalID
2113 * are NULL
2114 *
2115 * Returns a new document, do not initialize the DTD if not provided
2116 */
2117htmlDocPtr
2118htmlNewDocNoDtD(const xmlChar *URI, const xmlChar *ExternalID) {
2119 xmlDocPtr cur;
2120
2121 /*
2122 * Allocate a new document and fill the fields.
2123 */
2124 cur = (xmlDocPtr) xmlMalloc(sizeof(xmlDoc));
2125 if (cur == NULL) {
2126 htmlErrMemory(NULL, "HTML document creation failed\n");
2127 return(NULL);
2128 }
2129 memset(cur, 0, sizeof(xmlDoc));
2130
2131 cur->type = XML_HTML_DOCUMENT_NODE;
2132 cur->version = NULL;
2133 cur->intSubset = NULL;
2134 cur->doc = cur;
2135 cur->name = NULL;
2136 cur->children = NULL;
2137 cur->extSubset = NULL;
2138 cur->oldNs = NULL;
2139 cur->encoding = NULL;
2140 cur->standalone = 1;
2141 cur->compression = 0;
2142 cur->ids = NULL;
2143 cur->refs = NULL;
2144 cur->_private = NULL;
2145 cur->charset = XML_CHAR_ENCODING_UTF8;
2146 if ((ExternalID != NULL) ||
2147 (URI != NULL))
2148 xmlCreateIntSubset(cur, BAD_CAST "html", ExternalID, URI);
2149 return(cur);
2150}
2151
2152/**
2153 * htmlNewDoc:
2154 * @URI: URI for the dtd, or NULL
2155 * @ExternalID: the external ID of the DTD, or NULL
2156 *
2157 * Creates a new HTML document
2158 *
2159 * Returns a new document
2160 */
2161htmlDocPtr
2162htmlNewDoc(const xmlChar *URI, const xmlChar *ExternalID) {
2163 if ((URI == NULL) && (ExternalID == NULL))
2164 return(htmlNewDocNoDtD(
2165 BAD_CAST "http://www.w3.org/TR/REC-html40/loose.dtd",
2166 BAD_CAST "-//W3C//DTD HTML 4.0 Transitional//EN"));
2167
2168 return(htmlNewDocNoDtD(URI, ExternalID));
2169}
2170
2171
2172/************************************************************************
2173 * *
2174 * The parser itself *
2175 * Relates to http://www.w3.org/TR/html40 *
2176 * *
2177 ************************************************************************/
2178
2179/************************************************************************
2180 * *
2181 * The parser itself *
2182 * *
2183 ************************************************************************/
2184
2185static const xmlChar * htmlParseNameComplex(xmlParserCtxtPtr ctxt);
2186
2187/**
2188 * htmlParseHTMLName:
2189 * @ctxt: an HTML parser context
2190 *
2191 * parse an HTML tag or attribute name, note that we convert it to lowercase
2192 * since HTML names are not case-sensitive.
2193 *
2194 * Returns the Tag Name parsed or NULL
2195 */
2196
2197static const xmlChar *
2198htmlParseHTMLName(htmlParserCtxtPtr ctxt) {
2199 int i = 0;
2200 xmlChar loc[HTML_PARSER_BUFFER_SIZE];
2201
2202 if (!IS_ASCII_LETTER(CUR) && (CUR != '_') &&
2203 (CUR != ':')) return(NULL);
2204
2205 while ((i < HTML_PARSER_BUFFER_SIZE) &&
2206 ((IS_ASCII_LETTER(CUR)) || (IS_ASCII_DIGIT(CUR)) ||
2207 (CUR == ':') || (CUR == '-') || (CUR == '_'))) {
2208 if ((CUR >= 'A') && (CUR <= 'Z')) loc[i] = CUR + 0x20;
2209 else loc[i] = CUR;
2210 i++;
2211
2212 NEXT;
2213 }
2214
2215 return(xmlDictLookup(ctxt->dict, loc, i));
2216}
2217
2218
2219/**
2220 * htmlParseHTMLName_nonInvasive:
2221 * @ctxt: an HTML parser context
2222 *
2223 * parse an HTML tag or attribute name, note that we convert it to lowercase
2224 * since HTML names are not case-sensitive, this doesn't consume the data
2225 * from the stream, it's a look-ahead
2226 *
2227 * Returns the Tag Name parsed or NULL
2228 */
2229
2230static const xmlChar *
2231htmlParseHTMLName_nonInvasive(htmlParserCtxtPtr ctxt) {
2232 int i = 0;
2233 xmlChar loc[HTML_PARSER_BUFFER_SIZE];
2234
2235 if (!IS_ASCII_LETTER(NXT(1)) && (NXT(1) != '_') &&
2236 (NXT(1) != ':')) return(NULL);
2237
2238 while ((i < HTML_PARSER_BUFFER_SIZE) &&
2239 ((IS_ASCII_LETTER(NXT(1+i))) || (IS_ASCII_DIGIT(NXT(1+i))) ||
2240 (NXT(1+i) == ':') || (NXT(1+i) == '-') || (NXT(1+i) == '_'))) {
2241 if ((NXT(1+i) >= 'A') && (NXT(1+i) <= 'Z')) loc[i] = NXT(1+i) + 0x20;
2242 else loc[i] = NXT(1+i);
2243 i++;
2244 }
2245
2246 return(xmlDictLookup(ctxt->dict, loc, i));
2247}
2248
2249
2250/**
2251 * htmlParseName:
2252 * @ctxt: an HTML parser context
2253 *
2254 * parse an HTML name, this routine is case sensitive.
2255 *
2256 * Returns the Name parsed or NULL
2257 */
2258
2259static const xmlChar *
2260htmlParseName(htmlParserCtxtPtr ctxt) {
2261 const xmlChar *in;
2262 const xmlChar *ret;
2263 int count = 0;
2264
2265 GROW;
2266
2267 /*
2268 * Accelerator for simple ASCII names
2269 */
2270 in = ctxt->input->cur;
2271 if (((*in >= 0x61) && (*in <= 0x7A)) ||
2272 ((*in >= 0x41) && (*in <= 0x5A)) ||
2273 (*in == '_') || (*in == ':')) {
2274 in++;
2275 while (((*in >= 0x61) && (*in <= 0x7A)) ||
2276 ((*in >= 0x41) && (*in <= 0x5A)) ||
2277 ((*in >= 0x30) && (*in <= 0x39)) ||
2278 (*in == '_') || (*in == '-') ||
2279 (*in == ':') || (*in == '.'))
2280 in++;
2281 if ((*in > 0) && (*in < 0x80)) {
2282 count = in - ctxt->input->cur;
2283 ret = xmlDictLookup(ctxt->dict, ctxt->input->cur, count);
2284 ctxt->input->cur = in;
2285 ctxt->nbChars += count;
2286 ctxt->input->col += count;
2287 return(ret);
2288 }
2289 }
2290 return(htmlParseNameComplex(ctxt));
2291}
2292
2293static const xmlChar *
2294htmlParseNameComplex(xmlParserCtxtPtr ctxt) {
2295 int len = 0, l;
2296 int c;
2297 int count = 0;
2298
2299 /*
2300 * Handler for more complex cases
2301 */
2302 GROW;
2303 c = CUR_CHAR(l);
2304 if ((c == ' ') || (c == '>') || (c == '/') || /* accelerators */
2305 (!IS_LETTER(c) && (c != '_') &&
2306 (c != ':'))) {
2307 return(NULL);
2308 }
2309
2310 while ((c != ' ') && (c != '>') && (c != '/') && /* test bigname.xml */
2311 ((IS_LETTER(c)) || (IS_DIGIT(c)) ||
2312 (c == '.') || (c == '-') ||
2313 (c == '_') || (c == ':') ||
2314 (IS_COMBINING(c)) ||
2315 (IS_EXTENDER(c)))) {
2316 if (count++ > 100) {
2317 count = 0;
2318 GROW;
2319 }
2320 len += l;
2321 NEXTL(l);
2322 c = CUR_CHAR(l);
2323 }
2324 return(xmlDictLookup(ctxt->dict, ctxt->input->cur - len, len));
2325}
2326
2327
2328/**
2329 * htmlParseHTMLAttribute:
2330 * @ctxt: an HTML parser context
2331 * @stop: a char stop value
2332 *
2333 * parse an HTML attribute value till the stop (quote), if
2334 * stop is 0 then it stops at the first space
2335 *
2336 * Returns the attribute parsed or NULL
2337 */
2338
2339static xmlChar *
2340htmlParseHTMLAttribute(htmlParserCtxtPtr ctxt, const xmlChar stop) {
2341 xmlChar *buffer = NULL;
2342 int buffer_size = 0;
2343 xmlChar *out = NULL;
2344 const xmlChar *name = NULL;
2345 const xmlChar *cur = NULL;
2346 const htmlEntityDesc * ent;
2347
2348 /*
2349 * allocate a translation buffer.
2350 */
2351 buffer_size = HTML_PARSER_BUFFER_SIZE;
2352 buffer = (xmlChar *) xmlMallocAtomic(buffer_size * sizeof(xmlChar));
2353 if (buffer == NULL) {
2354 htmlErrMemory(ctxt, "buffer allocation failed\n");
2355 return(NULL);
2356 }
2357 out = buffer;
2358
2359 /*
2360 * Ok loop until we reach one of the ending chars
2361 */
2362 while ((CUR != 0) && (CUR != stop)) {
2363 if ((stop == 0) && (CUR == '>')) break;
2364 if ((stop == 0) && (IS_BLANK_CH(CUR))) break;
2365 if (CUR == '&') {
2366 if (NXT(1) == '#') {
2367 unsigned int c;
2368 int bits;
2369
2370 c = htmlParseCharRef(ctxt);
2371 if (c < 0x80)
2372 { *out++ = c; bits= -6; }
2373 else if (c < 0x800)
2374 { *out++ =((c >> 6) & 0x1F) | 0xC0; bits= 0; }
2375 else if (c < 0x10000)
2376 { *out++ =((c >> 12) & 0x0F) | 0xE0; bits= 6; }
2377 else
2378 { *out++ =((c >> 18) & 0x07) | 0xF0; bits= 12; }
2379
2380 for ( ; bits >= 0; bits-= 6) {
2381 *out++ = ((c >> bits) & 0x3F) | 0x80;
2382 }
2383
2384 if (out - buffer > buffer_size - 100) {
2385 int indx = out - buffer;
2386
2387 growBuffer(buffer);
2388 out = &buffer[indx];
2389 }
2390 } else {
2391 ent = htmlParseEntityRef(ctxt, &name);
2392 if (name == NULL) {
2393 *out++ = '&';
2394 if (out - buffer > buffer_size - 100) {
2395 int indx = out - buffer;
2396
2397 growBuffer(buffer);
2398 out = &buffer[indx];
2399 }
2400 } else if (ent == NULL) {
2401 *out++ = '&';
2402 cur = name;
2403 while (*cur != 0) {
2404 if (out - buffer > buffer_size - 100) {
2405 int indx = out - buffer;
2406
2407 growBuffer(buffer);
2408 out = &buffer[indx];
2409 }
2410 *out++ = *cur++;
2411 }
2412 } else {
2413 unsigned int c;
2414 int bits;
2415
2416 if (out - buffer > buffer_size - 100) {
2417 int indx = out - buffer;
2418
2419 growBuffer(buffer);
2420 out = &buffer[indx];
2421 }
2422 c = ent->value;
2423 if (c < 0x80)
2424 { *out++ = c; bits= -6; }
2425 else if (c < 0x800)
2426 { *out++ =((c >> 6) & 0x1F) | 0xC0; bits= 0; }
2427 else if (c < 0x10000)
2428 { *out++ =((c >> 12) & 0x0F) | 0xE0; bits= 6; }
2429 else
2430 { *out++ =((c >> 18) & 0x07) | 0xF0; bits= 12; }
2431
2432 for ( ; bits >= 0; bits-= 6) {
2433 *out++ = ((c >> bits) & 0x3F) | 0x80;
2434 }
2435 }
2436 }
2437 } else {
2438 unsigned int c;
2439 int bits, l;
2440
2441 if (out - buffer > buffer_size - 100) {
2442 int indx = out - buffer;
2443
2444 growBuffer(buffer);
2445 out = &buffer[indx];
2446 }
2447 c = CUR_CHAR(l);
2448 if (c < 0x80)
2449 { *out++ = c; bits= -6; }
2450 else if (c < 0x800)
2451 { *out++ =((c >> 6) & 0x1F) | 0xC0; bits= 0; }
2452 else if (c < 0x10000)
2453 { *out++ =((c >> 12) & 0x0F) | 0xE0; bits= 6; }
2454 else
2455 { *out++ =((c >> 18) & 0x07) | 0xF0; bits= 12; }
2456
2457 for ( ; bits >= 0; bits-= 6) {
2458 *out++ = ((c >> bits) & 0x3F) | 0x80;
2459 }
2460 NEXT;
2461 }
2462 }
2463 *out++ = 0;
2464 return(buffer);
2465}
2466
2467/**
2468 * htmlParseEntityRef:
2469 * @ctxt: an HTML parser context
2470 * @str: location to store the entity name
2471 *
2472 * parse an HTML ENTITY references
2473 *
2474 * [68] EntityRef ::= '&' Name ';'
2475 *
2476 * Returns the associated htmlEntityDescPtr if found, or NULL otherwise,
2477 * if non-NULL *str will have to be freed by the caller.
2478 */
2479const htmlEntityDesc *
2480htmlParseEntityRef(htmlParserCtxtPtr ctxt, const xmlChar **str) {
2481 const xmlChar *name;
2482 const htmlEntityDesc * ent = NULL;
2483
2484 if (str != NULL) *str = NULL;
2485 if ((ctxt == NULL) || (ctxt->input == NULL)) return(NULL);
2486
2487 if (CUR == '&') {
2488 NEXT;
2489 name = htmlParseName(ctxt);
2490 if (name == NULL) {
2491 htmlParseErr(ctxt, XML_ERR_NAME_REQUIRED,
2492 "htmlParseEntityRef: no name\n", NULL, NULL);
2493 } else {
2494 GROW;
2495 if (CUR == ';') {
2496 if (str != NULL)
2497 *str = name;
2498
2499 /*
2500 * Lookup the entity in the table.
2501 */
2502 ent = htmlEntityLookup(name);
2503 if (ent != NULL) /* OK that's ugly !!! */
2504 NEXT;
2505 } else {
2506 htmlParseErr(ctxt, XML_ERR_ENTITYREF_SEMICOL_MISSING,
2507 "htmlParseEntityRef: expecting ';'\n",
2508 NULL, NULL);
2509 if (str != NULL)
2510 *str = name;
2511 }
2512 }
2513 }
2514 return(ent);
2515}
2516
2517/**
2518 * htmlParseAttValue:
2519 * @ctxt: an HTML parser context
2520 *
2521 * parse a value for an attribute
2522 * Note: the parser won't do substitution of entities here, this
2523 * will be handled later in xmlStringGetNodeList, unless it was
2524 * asked for ctxt->replaceEntities != 0
2525 *
2526 * Returns the AttValue parsed or NULL.
2527 */
2528
2529static xmlChar *
2530htmlParseAttValue(htmlParserCtxtPtr ctxt) {
2531 xmlChar *ret = NULL;
2532
2533 if (CUR == '"') {
2534 NEXT;
2535 ret = htmlParseHTMLAttribute(ctxt, '"');
2536 if (CUR != '"') {
2537 htmlParseErr(ctxt, XML_ERR_ATTRIBUTE_NOT_FINISHED,
2538 "AttValue: \" expected\n", NULL, NULL);
2539 } else
2540 NEXT;
2541 } else if (CUR == '\'') {
2542 NEXT;
2543 ret = htmlParseHTMLAttribute(ctxt, '\'');
2544 if (CUR != '\'') {
2545 htmlParseErr(ctxt, XML_ERR_ATTRIBUTE_NOT_FINISHED,
2546 "AttValue: ' expected\n", NULL, NULL);
2547 } else
2548 NEXT;
2549 } else {
2550 /*
2551 * That's an HTMLism, the attribute value may not be quoted
2552 */
2553 ret = htmlParseHTMLAttribute(ctxt, 0);
2554 if (ret == NULL) {
2555 htmlParseErr(ctxt, XML_ERR_ATTRIBUTE_WITHOUT_VALUE,
2556 "AttValue: no value found\n", NULL, NULL);
2557 }
2558 }
2559 return(ret);
2560}
2561
2562/**
2563 * htmlParseSystemLiteral:
2564 * @ctxt: an HTML parser context
2565 *
2566 * parse an HTML Literal
2567 *
2568 * [11] SystemLiteral ::= ('"' [^"]* '"') | ("'" [^']* "'")
2569 *
2570 * Returns the SystemLiteral parsed or NULL
2571 */
2572
2573static xmlChar *
2574htmlParseSystemLiteral(htmlParserCtxtPtr ctxt) {
2575 const xmlChar *q;
2576 xmlChar *ret = NULL;
2577
2578 if (CUR == '"') {
2579 NEXT;
2580 q = CUR_PTR;
2581 while ((IS_CHAR_CH(CUR)) && (CUR != '"'))
2582 NEXT;
2583 if (!IS_CHAR_CH(CUR)) {
2584 htmlParseErr(ctxt, XML_ERR_LITERAL_NOT_FINISHED,
2585 "Unfinished SystemLiteral\n", NULL, NULL);
2586 } else {
2587 ret = xmlStrndup(q, CUR_PTR - q);
2588 NEXT;
2589 }
2590 } else if (CUR == '\'') {
2591 NEXT;
2592 q = CUR_PTR;
2593 while ((IS_CHAR_CH(CUR)) && (CUR != '\''))
2594 NEXT;
2595 if (!IS_CHAR_CH(CUR)) {
2596 htmlParseErr(ctxt, XML_ERR_LITERAL_NOT_FINISHED,
2597 "Unfinished SystemLiteral\n", NULL, NULL);
2598 } else {
2599 ret = xmlStrndup(q, CUR_PTR - q);
2600 NEXT;
2601 }
2602 } else {
2603 htmlParseErr(ctxt, XML_ERR_LITERAL_NOT_STARTED,
2604 " or ' expected\n", NULL, NULL);
2605 }
2606
2607 return(ret);
2608}
2609
2610/**
2611 * htmlParsePubidLiteral:
2612 * @ctxt: an HTML parser context
2613 *
2614 * parse an HTML public literal
2615 *
2616 * [12] PubidLiteral ::= '"' PubidChar* '"' | "'" (PubidChar - "'")* "'"
2617 *
2618 * Returns the PubidLiteral parsed or NULL.
2619 */
2620
2621static xmlChar *
2622htmlParsePubidLiteral(htmlParserCtxtPtr ctxt) {
2623 const xmlChar *q;
2624 xmlChar *ret = NULL;
2625 /*
2626 * Name ::= (Letter | '_') (NameChar)*
2627 */
2628 if (CUR == '"') {
2629 NEXT;
2630 q = CUR_PTR;
2631 while (IS_PUBIDCHAR_CH(CUR)) NEXT;
2632 if (CUR != '"') {
2633 htmlParseErr(ctxt, XML_ERR_LITERAL_NOT_FINISHED,
2634 "Unfinished PubidLiteral\n", NULL, NULL);
2635 } else {
2636 ret = xmlStrndup(q, CUR_PTR - q);
2637 NEXT;
2638 }
2639 } else if (CUR == '\'') {
2640 NEXT;
2641 q = CUR_PTR;
2642 while ((IS_PUBIDCHAR_CH(CUR)) && (CUR != '\''))
2643 NEXT;
2644 if (CUR != '\'') {
2645 htmlParseErr(ctxt, XML_ERR_LITERAL_NOT_FINISHED,
2646 "Unfinished PubidLiteral\n", NULL, NULL);
2647 } else {
2648 ret = xmlStrndup(q, CUR_PTR - q);
2649 NEXT;
2650 }
2651 } else {
2652 htmlParseErr(ctxt, XML_ERR_LITERAL_NOT_STARTED,
2653 "PubidLiteral \" or ' expected\n", NULL, NULL);
2654 }
2655
2656 return(ret);
2657}
2658
2659/**
2660 * htmlParseScript:
2661 * @ctxt: an HTML parser context
2662 *
2663 * parse the content of an HTML SCRIPT or STYLE element
2664 * http://www.w3.org/TR/html4/sgml/dtd.html#Script
2665 * http://www.w3.org/TR/html4/sgml/dtd.html#StyleSheet
2666 * http://www.w3.org/TR/html4/types.html#type-script
2667 * http://www.w3.org/TR/html4/types.html#h-6.15
2668 * http://www.w3.org/TR/html4/appendix/notes.html#h-B.3.2.1
2669 *
2670 * Script data ( %Script; in the DTD) can be the content of the SCRIPT
2671 * element and the value of intrinsic event attributes. User agents must
2672 * not evaluate script data as HTML markup but instead must pass it on as
2673 * data to a script engine.
2674 * NOTES:
2675 * - The content is passed like CDATA
2676 * - the attributes for style and scripting "onXXX" are also described
2677 * as CDATA but SGML allows entities references in attributes so their
2678 * processing is identical as other attributes
2679 */
2680static void
2681htmlParseScript(htmlParserCtxtPtr ctxt) {
2682 xmlChar buf[HTML_PARSER_BIG_BUFFER_SIZE + 5];
2683 int nbchar = 0;
2684 int cur,l;
2685
2686 SHRINK;
2687 cur = CUR_CHAR(l);
2688 while (IS_CHAR_CH(cur)) {
2689 if ((cur == '<') && (NXT(1) == '/')) {
2690 /*
2691 * One should break here, the specification is clear:
2692 * Authors should therefore escape "</" within the content.
2693 * Escape mechanisms are specific to each scripting or
2694 * style sheet language.
2695 *
2696 * In recovery mode, only break if end tag match the
2697 * current tag, effectively ignoring all tags inside the
2698 * script/style block and treating the entire block as
2699 * CDATA.
2700 */
2701 if (ctxt->recovery) {
2702 if (xmlStrncasecmp(ctxt->name, ctxt->input->cur+2,
2703 xmlStrlen(ctxt->name)) == 0)
2704 {
2705 break; /* while */
2706 } else {
2707 htmlParseErr(ctxt, XML_ERR_TAG_NAME_MISMATCH,
2708 "Element %s embeds close tag\n",
2709 ctxt->name, NULL);
2710 }
2711 } else {
2712 if (((NXT(2) >= 'A') && (NXT(2) <= 'Z')) ||
2713 ((NXT(2) >= 'a') && (NXT(2) <= 'z')))
2714 {
2715 break; /* while */
2716 }
2717 }
2718 }
2719 COPY_BUF(l,buf,nbchar,cur);
2720 if (nbchar >= HTML_PARSER_BIG_BUFFER_SIZE) {
2721 if (ctxt->sax->cdataBlock!= NULL) {
2722 /*
2723 * Insert as CDATA, which is the same as HTML_PRESERVE_NODE
2724 */
2725 ctxt->sax->cdataBlock(ctxt->userData, buf, nbchar);
2726 } else if (ctxt->sax->characters != NULL) {
2727 ctxt->sax->characters(ctxt->userData, buf, nbchar);
2728 }
2729 nbchar = 0;
2730 }
2731 GROW;
2732 NEXTL(l);
2733 cur = CUR_CHAR(l);
2734 }
2735
2736 if ((!(IS_CHAR_CH(cur))) && (!((cur == 0) && (ctxt->progressive)))) {
2737 htmlParseErrInt(ctxt, XML_ERR_INVALID_CHAR,
2738 "Invalid char in CDATA 0x%X\n", cur);
2739 NEXT;
2740 }
2741
2742 if ((nbchar != 0) && (ctxt->sax != NULL) && (!ctxt->disableSAX)) {
2743 if (ctxt->sax->cdataBlock!= NULL) {
2744 /*
2745 * Insert as CDATA, which is the same as HTML_PRESERVE_NODE
2746 */
2747 ctxt->sax->cdataBlock(ctxt->userData, buf, nbchar);
2748 } else if (ctxt->sax->characters != NULL) {
2749 ctxt->sax->characters(ctxt->userData, buf, nbchar);
2750 }
2751 }
2752}
2753
2754
2755/**
2756 * htmlParseCharData:
2757 * @ctxt: an HTML parser context
2758 *
2759 * parse a CharData section.
2760 * if we are within a CDATA section ']]>' marks an end of section.
2761 *
2762 * [14] CharData ::= [^<&]* - ([^<&]* ']]>' [^<&]*)
2763 */
2764
2765static void
2766htmlParseCharData(htmlParserCtxtPtr ctxt) {
2767 xmlChar buf[HTML_PARSER_BIG_BUFFER_SIZE + 5];
2768 int nbchar = 0;
2769 int cur, l;
2770
2771 SHRINK;
2772 cur = CUR_CHAR(l);
2773 while (((cur != '<') || (ctxt->token == '<')) &&
2774 ((cur != '&') || (ctxt->token == '&')) &&
2775 (IS_CHAR(cur))) {
2776 COPY_BUF(l,buf,nbchar,cur);
2777 if (nbchar >= HTML_PARSER_BIG_BUFFER_SIZE) {
2778 /*
2779 * Ok the segment is to be consumed as chars.
2780 */
2781 if ((ctxt->sax != NULL) && (!ctxt->disableSAX)) {
2782 if (areBlanks(ctxt, buf, nbchar)) {
2783 if (ctxt->sax->ignorableWhitespace != NULL)
2784 ctxt->sax->ignorableWhitespace(ctxt->userData,
2785 buf, nbchar);
2786 } else {
2787 htmlCheckParagraph(ctxt);
2788 if (ctxt->sax->characters != NULL)
2789 ctxt->sax->characters(ctxt->userData, buf, nbchar);
2790 }
2791 }
2792 nbchar = 0;
2793 }
2794 NEXTL(l);
2795 cur = CUR_CHAR(l);
2796 if (cur == 0) {
2797 SHRINK;
2798 GROW;
2799 cur = CUR_CHAR(l);
2800 }
2801 }
2802 if (nbchar != 0) {
2803 buf[nbchar] = 0;
2804
2805 /*
2806 * Ok the segment is to be consumed as chars.
2807 */
2808 if ((ctxt->sax != NULL) && (!ctxt->disableSAX)) {
2809 if (areBlanks(ctxt, buf, nbchar)) {
2810 if (ctxt->sax->ignorableWhitespace != NULL)
2811 ctxt->sax->ignorableWhitespace(ctxt->userData, buf, nbchar);
2812 } else {
2813 htmlCheckParagraph(ctxt);
2814 if (ctxt->sax->characters != NULL)
2815 ctxt->sax->characters(ctxt->userData, buf, nbchar);
2816 }
2817 }
2818 } else {
2819 /*
2820 * Loop detection
2821 */
2822 if (cur == 0)
2823 ctxt->instate = XML_PARSER_EOF;
2824 }
2825}
2826
2827/**
2828 * htmlParseExternalID:
2829 * @ctxt: an HTML parser context
2830 * @publicID: a xmlChar** receiving PubidLiteral
2831 *
2832 * Parse an External ID or a Public ID
2833 *
2834 * [75] ExternalID ::= 'SYSTEM' S SystemLiteral
2835 * | 'PUBLIC' S PubidLiteral S SystemLiteral
2836 *
2837 * [83] PublicID ::= 'PUBLIC' S PubidLiteral
2838 *
2839 * Returns the function returns SystemLiteral and in the second
2840 * case publicID receives PubidLiteral, is strict is off
2841 * it is possible to return NULL and have publicID set.
2842 */
2843
2844static xmlChar *
2845htmlParseExternalID(htmlParserCtxtPtr ctxt, xmlChar **publicID) {
2846 xmlChar *URI = NULL;
2847
2848 if ((UPPER == 'S') && (UPP(1) == 'Y') &&
2849 (UPP(2) == 'S') && (UPP(3) == 'T') &&
2850 (UPP(4) == 'E') && (UPP(5) == 'M')) {
2851 SKIP(6);
2852 if (!IS_BLANK_CH(CUR)) {
2853 htmlParseErr(ctxt, XML_ERR_SPACE_REQUIRED,
2854 "Space required after 'SYSTEM'\n", NULL, NULL);
2855 }
2856 SKIP_BLANKS;
2857 URI = htmlParseSystemLiteral(ctxt);
2858 if (URI == NULL) {
2859 htmlParseErr(ctxt, XML_ERR_URI_REQUIRED,
2860 "htmlParseExternalID: SYSTEM, no URI\n", NULL, NULL);
2861 }
2862 } else if ((UPPER == 'P') && (UPP(1) == 'U') &&
2863 (UPP(2) == 'B') && (UPP(3) == 'L') &&
2864 (UPP(4) == 'I') && (UPP(5) == 'C')) {
2865 SKIP(6);
2866 if (!IS_BLANK_CH(CUR)) {
2867 htmlParseErr(ctxt, XML_ERR_SPACE_REQUIRED,
2868 "Space required after 'PUBLIC'\n", NULL, NULL);
2869 }
2870 SKIP_BLANKS;
2871 *publicID = htmlParsePubidLiteral(ctxt);
2872 if (*publicID == NULL) {
2873 htmlParseErr(ctxt, XML_ERR_PUBID_REQUIRED,
2874 "htmlParseExternalID: PUBLIC, no Public Identifier\n",
2875 NULL, NULL);
2876 }
2877 SKIP_BLANKS;
2878 if ((CUR == '"') || (CUR == '\'')) {
2879 URI = htmlParseSystemLiteral(ctxt);
2880 }
2881 }
2882 return(URI);
2883}
2884
2885/**
2886 * xmlParsePI:
2887 * @ctxt: an XML parser context
2888 *
2889 * parse an XML Processing Instruction.
2890 *
2891 * [16] PI ::= '<?' PITarget (S (Char* - (Char* '?>' Char*)))? '?>'
2892 */
2893static void
2894htmlParsePI(htmlParserCtxtPtr ctxt) {
2895 xmlChar *buf = NULL;
2896 int len = 0;
2897 int size = HTML_PARSER_BUFFER_SIZE;
2898 int cur, l;
2899 const xmlChar *target;
2900 xmlParserInputState state;
2901 int count = 0;
2902
2903 if ((RAW == '<') && (NXT(1) == '?')) {
2904 state = ctxt->instate;
2905 ctxt->instate = XML_PARSER_PI;
2906 /*
2907 * this is a Processing Instruction.
2908 */
2909 SKIP(2);
2910 SHRINK;
2911
2912 /*
2913 * Parse the target name and check for special support like
2914 * namespace.
2915 */
2916 target = htmlParseName(ctxt);
2917 if (target != NULL) {
2918 if (RAW == '>') {
2919 SKIP(1);
2920
2921 /*
2922 * SAX: PI detected.
2923 */
2924 if ((ctxt->sax) && (!ctxt->disableSAX) &&
2925 (ctxt->sax->processingInstruction != NULL))
2926 ctxt->sax->processingInstruction(ctxt->userData,
2927 target, NULL);
2928 ctxt->instate = state;
2929 return;
2930 }
2931 buf = (xmlChar *) xmlMallocAtomic(size * sizeof(xmlChar));
2932 if (buf == NULL) {
2933 htmlErrMemory(ctxt, NULL);
2934 ctxt->instate = state;
2935 return;
2936 }
2937 cur = CUR;
2938 if (!IS_BLANK(cur)) {
2939 htmlParseErr(ctxt, XML_ERR_SPACE_REQUIRED,
2940 "ParsePI: PI %s space expected\n", target, NULL);
2941 }
2942 SKIP_BLANKS;
2943 cur = CUR_CHAR(l);
2944 while (IS_CHAR(cur) && (cur != '>')) {
2945 if (len + 5 >= size) {
2946 xmlChar *tmp;
2947
2948 size *= 2;
2949 tmp = (xmlChar *) xmlRealloc(buf, size * sizeof(xmlChar));
2950 if (tmp == NULL) {
2951 htmlErrMemory(ctxt, NULL);
2952 xmlFree(buf);
2953 ctxt->instate = state;
2954 return;
2955 }
2956 buf = tmp;
2957 }
2958 count++;
2959 if (count > 50) {
2960 GROW;
2961 count = 0;
2962 }
2963 COPY_BUF(l,buf,len,cur);
2964 NEXTL(l);
2965 cur = CUR_CHAR(l);
2966 if (cur == 0) {
2967 SHRINK;
2968 GROW;
2969 cur = CUR_CHAR(l);
2970 }
2971 }
2972 buf[len] = 0;
2973 if (cur != '>') {
2974 htmlParseErr(ctxt, XML_ERR_PI_NOT_FINISHED,
2975 "ParsePI: PI %s never end ...\n", target, NULL);
2976 } else {
2977 SKIP(1);
2978
2979 /*
2980 * SAX: PI detected.
2981 */
2982 if ((ctxt->sax) && (!ctxt->disableSAX) &&
2983 (ctxt->sax->processingInstruction != NULL))
2984 ctxt->sax->processingInstruction(ctxt->userData,
2985 target, buf);
2986 }
2987 xmlFree(buf);
2988 } else {
2989 htmlParseErr(ctxt, XML_ERR_PI_NOT_STARTED,
2990 "PI is not started correctly", NULL, NULL);
2991 }
2992 ctxt->instate = state;
2993 }
2994}
2995
2996/**
2997 * htmlParseComment:
2998 * @ctxt: an HTML parser context
2999 *
3000 * Parse an XML (SGML) comment <!-- .... -->
3001 *
3002 * [15] Comment ::= '<!--' ((Char - '-') | ('-' (Char - '-')))* '-->'
3003 */
3004static void
3005htmlParseComment(htmlParserCtxtPtr ctxt) {
3006 xmlChar *buf = NULL;
3007 int len;
3008 int size = HTML_PARSER_BUFFER_SIZE;
3009 int q, ql;
3010 int r, rl;
3011 int cur, l;
3012 xmlParserInputState state;
3013
3014 /*
3015 * Check that there is a comment right here.
3016 */
3017 if ((RAW != '<') || (NXT(1) != '!') ||
3018 (NXT(2) != '-') || (NXT(3) != '-')) return;
3019
3020 state = ctxt->instate;
3021 ctxt->instate = XML_PARSER_COMMENT;
3022 SHRINK;
3023 SKIP(4);
3024 buf = (xmlChar *) xmlMallocAtomic(size * sizeof(xmlChar));
3025 if (buf == NULL) {
3026 htmlErrMemory(ctxt, "buffer allocation failed\n");
3027 ctxt->instate = state;
3028 return;
3029 }
3030 q = CUR_CHAR(ql);
3031 NEXTL(ql);
3032 r = CUR_CHAR(rl);
3033 NEXTL(rl);
3034 cur = CUR_CHAR(l);
3035 len = 0;
3036 while (IS_CHAR(cur) &&
3037 ((cur != '>') ||
3038 (r != '-') || (q != '-'))) {
3039 if (len + 5 >= size) {
3040 xmlChar *tmp;
3041
3042 size *= 2;
3043 tmp = (xmlChar *) xmlRealloc(buf, size * sizeof(xmlChar));
3044 if (tmp == NULL) {
3045 xmlFree(buf);
3046 htmlErrMemory(ctxt, "growing buffer failed\n");
3047 ctxt->instate = state;
3048 return;
3049 }
3050 buf = tmp;
3051 }
3052 COPY_BUF(ql,buf,len,q);
3053 q = r;
3054 ql = rl;
3055 r = cur;
3056 rl = l;
3057 NEXTL(l);
3058 cur = CUR_CHAR(l);
3059 if (cur == 0) {
3060 SHRINK;
3061 GROW;
3062 cur = CUR_CHAR(l);
3063 }
3064 }
3065 buf[len] = 0;
3066 if (!IS_CHAR(cur)) {
3067 htmlParseErr(ctxt, XML_ERR_COMMENT_NOT_FINISHED,
3068 "Comment not terminated \n<!--%.50s\n", buf, NULL);
3069 xmlFree(buf);
3070 } else {
3071 NEXT;
3072 if ((ctxt->sax != NULL) && (ctxt->sax->comment != NULL) &&
3073 (!ctxt->disableSAX))
3074 ctxt->sax->comment(ctxt->userData, buf);
3075 xmlFree(buf);
3076 }
3077 ctxt->instate = state;
3078}
3079
3080/**
3081 * htmlParseCharRef:
3082 * @ctxt: an HTML parser context
3083 *
3084 * parse Reference declarations
3085 *
3086 * [66] CharRef ::= '&#' [0-9]+ ';' |
3087 * '&#x' [0-9a-fA-F]+ ';'
3088 *
3089 * Returns the value parsed (as an int)
3090 */
3091int
3092htmlParseCharRef(htmlParserCtxtPtr ctxt) {
3093 int val = 0;
3094
3095 if ((ctxt == NULL) || (ctxt->input == NULL)) {
3096 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
3097 "htmlParseCharRef: context error\n",
3098 NULL, NULL);
3099 return(0);
3100 }
3101 if ((CUR == '&') && (NXT(1) == '#') &&
3102 ((NXT(2) == 'x') || NXT(2) == 'X')) {
3103 SKIP(3);
3104 while (CUR != ';') {
3105 if ((CUR >= '0') && (CUR <= '9'))
3106 val = val * 16 + (CUR - '0');
3107 else if ((CUR >= 'a') && (CUR <= 'f'))
3108 val = val * 16 + (CUR - 'a') + 10;
3109 else if ((CUR >= 'A') && (CUR <= 'F'))
3110 val = val * 16 + (CUR - 'A') + 10;
3111 else {
3112 htmlParseErr(ctxt, XML_ERR_INVALID_HEX_CHARREF,
3113 "htmlParseCharRef: invalid hexadecimal value\n",
3114 NULL, NULL);
3115 return(0);
3116 }
3117 NEXT;
3118 }
3119 if (CUR == ';')
3120 NEXT;
3121 } else if ((CUR == '&') && (NXT(1) == '#')) {
3122 SKIP(2);
3123 while (CUR != ';') {
3124 if ((CUR >= '0') && (CUR <= '9'))
3125 val = val * 10 + (CUR - '0');
3126 else {
3127 htmlParseErr(ctxt, XML_ERR_INVALID_DEC_CHARREF,
3128 "htmlParseCharRef: invalid decimal value\n",
3129 NULL, NULL);
3130 return(0);
3131 }
3132 NEXT;
3133 }
3134 if (CUR == ';')
3135 NEXT;
3136 } else {
3137 htmlParseErr(ctxt, XML_ERR_INVALID_CHARREF,
3138 "htmlParseCharRef: invalid value\n", NULL, NULL);
3139 }
3140 /*
3141 * Check the value IS_CHAR ...
3142 */
3143 if (IS_CHAR(val)) {
3144 return(val);
3145 } else {
3146 htmlParseErrInt(ctxt, XML_ERR_INVALID_CHAR,
3147 "htmlParseCharRef: invalid xmlChar value %d\n",
3148 val);
3149 }
3150 return(0);
3151}
3152
3153
3154/**
3155 * htmlParseDocTypeDecl:
3156 * @ctxt: an HTML parser context
3157 *
3158 * parse a DOCTYPE declaration
3159 *
3160 * [28] doctypedecl ::= '<!DOCTYPE' S Name (S ExternalID)? S?
3161 * ('[' (markupdecl | PEReference | S)* ']' S?)? '>'
3162 */
3163
3164static void
3165htmlParseDocTypeDecl(htmlParserCtxtPtr ctxt) {
3166 const xmlChar *name;
3167 xmlChar *ExternalID = NULL;
3168 xmlChar *URI = NULL;
3169
3170 /*
3171 * We know that '<!DOCTYPE' has been detected.
3172 */
3173 SKIP(9);
3174
3175 SKIP_BLANKS;
3176
3177 /*
3178 * Parse the DOCTYPE name.
3179 */
3180 name = htmlParseName(ctxt);
3181 if (name == NULL) {
3182 htmlParseErr(ctxt, XML_ERR_NAME_REQUIRED,
3183 "htmlParseDocTypeDecl : no DOCTYPE name !\n",
3184 NULL, NULL);
3185 }
3186 /*
3187 * Check that upper(name) == "HTML" !!!!!!!!!!!!!
3188 */
3189
3190 SKIP_BLANKS;
3191
3192 /*
3193 * Check for SystemID and ExternalID
3194 */
3195 URI = htmlParseExternalID(ctxt, &ExternalID);
3196 SKIP_BLANKS;
3197
3198 /*
3199 * We should be at the end of the DOCTYPE declaration.
3200 */
3201 if (CUR != '>') {
3202 htmlParseErr(ctxt, XML_ERR_DOCTYPE_NOT_FINISHED,
3203 "DOCTYPE improperly terminated\n", NULL, NULL);
3204 /* We shouldn't try to resynchronize ... */
3205 }
3206 NEXT;
3207
3208 /*
3209 * Create or update the document accordingly to the DOCTYPE
3210 */
3211 if ((ctxt->sax != NULL) && (ctxt->sax->internalSubset != NULL) &&
3212 (!ctxt->disableSAX))
3213 ctxt->sax->internalSubset(ctxt->userData, name, ExternalID, URI);
3214
3215 /*
3216 * Cleanup, since we don't use all those identifiers
3217 */
3218 if (URI != NULL) xmlFree(URI);
3219 if (ExternalID != NULL) xmlFree(ExternalID);
3220}
3221
3222/**
3223 * htmlParseAttribute:
3224 * @ctxt: an HTML parser context
3225 * @value: a xmlChar ** used to store the value of the attribute
3226 *
3227 * parse an attribute
3228 *
3229 * [41] Attribute ::= Name Eq AttValue
3230 *
3231 * [25] Eq ::= S? '=' S?
3232 *
3233 * With namespace:
3234 *
3235 * [NS 11] Attribute ::= QName Eq AttValue
3236 *
3237 * Also the case QName == xmlns:??? is handled independently as a namespace
3238 * definition.
3239 *
3240 * Returns the attribute name, and the value in *value.
3241 */
3242
3243static const xmlChar *
3244htmlParseAttribute(htmlParserCtxtPtr ctxt, xmlChar **value) {
3245 const xmlChar *name;
3246 xmlChar *val = NULL;
3247
3248 *value = NULL;
3249 name = htmlParseHTMLName(ctxt);
3250 if (name == NULL) {
3251 htmlParseErr(ctxt, XML_ERR_NAME_REQUIRED,
3252 "error parsing attribute name\n", NULL, NULL);
3253 return(NULL);
3254 }
3255
3256 /*
3257 * read the value
3258 */
3259 SKIP_BLANKS;
3260 if (CUR == '=') {
3261 NEXT;
3262 SKIP_BLANKS;
3263 val = htmlParseAttValue(ctxt);
3264 } else if (htmlIsBooleanAttr(name)) {
3265 /*
3266 * assume a minimized attribute
3267 */
3268 val = xmlStrdup(name);
3269 }
3270
3271 *value = val;
3272 return(name);
3273}
3274
3275/**
3276 * htmlCheckEncoding:
3277 * @ctxt: an HTML parser context
3278 * @attvalue: the attribute value
3279 *
3280 * Checks an http-equiv attribute from a Meta tag to detect
3281 * the encoding
3282 * If a new encoding is detected the parser is switched to decode
3283 * it and pass UTF8
3284 */
3285static void
3286htmlCheckEncoding(htmlParserCtxtPtr ctxt, const xmlChar *attvalue) {
3287 const xmlChar *encoding;
3288
3289 if ((ctxt == NULL) || (attvalue == NULL))
3290 return;
3291
3292 /* do not change encoding */
3293 if (ctxt->input->encoding != NULL)
3294 return;
3295
3296 encoding = xmlStrcasestr(attvalue, BAD_CAST"charset=");
3297 if (encoding != NULL) {
3298 encoding += 8;
3299 } else {
3300 encoding = xmlStrcasestr(attvalue, BAD_CAST"charset =");
3301 if (encoding != NULL)
3302 encoding += 9;
3303 }
3304 if (encoding != NULL) {
3305 xmlCharEncoding enc;
3306 xmlCharEncodingHandlerPtr handler;
3307
3308 while ((*encoding == ' ') || (*encoding == '\t')) encoding++;
3309
3310 if (ctxt->input->encoding != NULL)
3311 xmlFree((xmlChar *) ctxt->input->encoding);
3312 ctxt->input->encoding = xmlStrdup(encoding);
3313
3314 enc = xmlParseCharEncoding((const char *) encoding);
3315 /*
3316 * registered set of known encodings
3317 */
3318 if (enc != XML_CHAR_ENCODING_ERROR) {
3319 if (((enc == XML_CHAR_ENCODING_UTF16LE) ||
3320 (enc == XML_CHAR_ENCODING_UTF16BE) ||
3321 (enc == XML_CHAR_ENCODING_UCS4LE) ||
3322 (enc == XML_CHAR_ENCODING_UCS4BE)) &&
3323 (ctxt->input->buf != NULL) &&
3324 (ctxt->input->buf->encoder == NULL)) {
3325 htmlParseErr(ctxt, XML_ERR_INVALID_ENCODING,
3326 "htmlCheckEncoding: wrong encoding meta\n",
3327 NULL, NULL);
3328 } else {
3329 xmlSwitchEncoding(ctxt, enc);
3330 }
3331 ctxt->charset = XML_CHAR_ENCODING_UTF8;
3332 } else {
3333 /*
3334 * fallback for unknown encodings
3335 */
3336 handler = xmlFindCharEncodingHandler((const char *) encoding);
3337 if (handler != NULL) {
3338 xmlSwitchToEncoding(ctxt, handler);
3339 ctxt->charset = XML_CHAR_ENCODING_UTF8;
3340 } else {
3341 ctxt->errNo = XML_ERR_UNSUPPORTED_ENCODING;
3342 }
3343 }
3344
3345 if ((ctxt->input->buf != NULL) &&
3346 (ctxt->input->buf->encoder != NULL) &&
3347 (ctxt->input->buf->raw != NULL) &&
3348 (ctxt->input->buf->buffer != NULL)) {
3349 int nbchars;
3350 int processed;
3351
3352 /*
3353 * convert as much as possible to the parser reading buffer.
3354 */
3355 processed = ctxt->input->cur - ctxt->input->base;
3356 xmlBufferShrink(ctxt->input->buf->buffer, processed);
3357 nbchars = xmlCharEncInFunc(ctxt->input->buf->encoder,
3358 ctxt->input->buf->buffer,
3359 ctxt->input->buf->raw);
3360 if (nbchars < 0) {
3361 htmlParseErr(ctxt, XML_ERR_INVALID_ENCODING,
3362 "htmlCheckEncoding: encoder error\n",
3363 NULL, NULL);
3364 }
3365 ctxt->input->base =
3366 ctxt->input->cur = ctxt->input->buf->buffer->content;
3367 }
3368 }
3369}
3370
3371/**
3372 * htmlCheckMeta:
3373 * @ctxt: an HTML parser context
3374 * @atts: the attributes values
3375 *
3376 * Checks an attributes from a Meta tag
3377 */
3378static void
3379htmlCheckMeta(htmlParserCtxtPtr ctxt, const xmlChar **atts) {
3380 int i;
3381 const xmlChar *att, *value;
3382 int http = 0;
3383 const xmlChar *content = NULL;
3384
3385 if ((ctxt == NULL) || (atts == NULL))
3386 return;
3387
3388 i = 0;
3389 att = atts[i++];
3390 while (att != NULL) {
3391 value = atts[i++];
3392 if ((value != NULL) && (!xmlStrcasecmp(att, BAD_CAST"http-equiv"))
3393 && (!xmlStrcasecmp(value, BAD_CAST"Content-Type")))
3394 http = 1;
3395 else if ((value != NULL) && (!xmlStrcasecmp(att, BAD_CAST"content")))
3396 content = value;
3397 att = atts[i++];
3398 }
3399 if ((http) && (content != NULL))
3400 htmlCheckEncoding(ctxt, content);
3401
3402}
3403
3404/**
3405 * htmlParseStartTag:
3406 * @ctxt: an HTML parser context
3407 *
3408 * parse a start of tag either for rule element or
3409 * EmptyElement. In both case we don't parse the tag closing chars.
3410 *
3411 * [40] STag ::= '<' Name (S Attribute)* S? '>'
3412 *
3413 * [44] EmptyElemTag ::= '<' Name (S Attribute)* S? '/>'
3414 *
3415 * With namespace:
3416 *
3417 * [NS 8] STag ::= '<' QName (S Attribute)* S? '>'
3418 *
3419 * [NS 10] EmptyElement ::= '<' QName (S Attribute)* S? '/>'
3420 *
3421 * Returns 0 in case of success and -1 in case of error.
3422 */
3423
3424static int
3425htmlParseStartTag(htmlParserCtxtPtr ctxt) {
3426 const xmlChar *name;
3427 const xmlChar *attname;
3428 xmlChar *attvalue;
3429 const xmlChar **atts;
3430 int nbatts = 0;
3431 int maxatts;
3432 int meta = 0;
3433 int i;
3434
3435 if ((ctxt == NULL) || (ctxt->input == NULL)) {
3436 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
3437 "htmlParseStartTag: context error\n", NULL, NULL);
3438 return -1;
3439 }
3440 if (CUR != '<') return -1;
3441 NEXT;
3442
3443 atts = ctxt->atts;
3444 maxatts = ctxt->maxatts;
3445
3446 GROW;
3447 name = htmlParseHTMLName(ctxt);
3448 if (name == NULL) {
3449 htmlParseErr(ctxt, XML_ERR_NAME_REQUIRED,
3450 "htmlParseStartTag: invalid element name\n",
3451 NULL, NULL);
3452 /* Dump the bogus tag like browsers do */
3453 while ((IS_CHAR_CH(CUR)) && (CUR != '>'))
3454 NEXT;
3455 return -1;
3456 }
3457 if (xmlStrEqual(name, BAD_CAST"meta"))
3458 meta = 1;
3459
3460 /*
3461 * Check for auto-closure of HTML elements.
3462 */
3463 htmlAutoClose(ctxt, name);
3464
3465 /*
3466 * Check for implied HTML elements.
3467 */
3468 htmlCheckImplied(ctxt, name);
3469
3470 /*
3471 * Avoid html at any level > 0, head at any level != 1
3472 * or any attempt to recurse body
3473 */
3474 if ((ctxt->nameNr > 0) && (xmlStrEqual(name, BAD_CAST"html"))) {
3475 htmlParseErr(ctxt, XML_HTML_STRUCURE_ERROR,
3476 "htmlParseStartTag: misplaced <html> tag\n",
3477 name, NULL);
3478 return 0;
3479 }
3480 if ((ctxt->nameNr != 1) &&
3481 (xmlStrEqual(name, BAD_CAST"head"))) {
3482 htmlParseErr(ctxt, XML_HTML_STRUCURE_ERROR,
3483 "htmlParseStartTag: misplaced <head> tag\n",
3484 name, NULL);
3485 return 0;
3486 }
3487 if (xmlStrEqual(name, BAD_CAST"body")) {
3488 int indx;
3489 for (indx = 0;indx < ctxt->nameNr;indx++) {
3490 if (xmlStrEqual(ctxt->nameTab[indx], BAD_CAST"body")) {
3491 htmlParseErr(ctxt, XML_HTML_STRUCURE_ERROR,
3492 "htmlParseStartTag: misplaced <body> tag\n",
3493 name, NULL);
3494 while ((IS_CHAR_CH(CUR)) && (CUR != '>'))
3495 NEXT;
3496 return 0;
3497 }
3498 }
3499 }
3500
3501 /*
3502 * Now parse the attributes, it ends up with the ending
3503 *
3504 * (S Attribute)* S?
3505 */
3506 SKIP_BLANKS;
3507 while ((IS_CHAR_CH(CUR)) &&
3508 (CUR != '>') &&
3509 ((CUR != '/') || (NXT(1) != '>'))) {
3510 long cons = ctxt->nbChars;
3511
3512 GROW;
3513 attname = htmlParseAttribute(ctxt, &attvalue);
3514 if (attname != NULL) {
3515
3516 /*
3517 * Well formedness requires at most one declaration of an attribute
3518 */
3519 for (i = 0; i < nbatts;i += 2) {
3520 if (xmlStrEqual(atts[i], attname)) {
3521 htmlParseErr(ctxt, XML_ERR_ATTRIBUTE_REDEFINED,
3522 "Attribute %s redefined\n", attname, NULL);
3523 if (attvalue != NULL)
3524 xmlFree(attvalue);
3525 goto failed;
3526 }
3527 }
3528
3529 /*
3530 * Add the pair to atts
3531 */
3532 if (atts == NULL) {
3533 maxatts = 22; /* allow for 10 attrs by default */
3534 atts = (const xmlChar **)
3535 xmlMalloc(maxatts * sizeof(xmlChar *));
3536 if (atts == NULL) {
3537 htmlErrMemory(ctxt, NULL);
3538 if (attvalue != NULL)
3539 xmlFree(attvalue);
3540 goto failed;
3541 }
3542 ctxt->atts = atts;
3543 ctxt->maxatts = maxatts;
3544 } else if (nbatts + 4 > maxatts) {
3545 const xmlChar **n;
3546
3547 maxatts *= 2;
3548 n = (const xmlChar **) xmlRealloc((void *) atts,
3549 maxatts * sizeof(const xmlChar *));
3550 if (n == NULL) {
3551 htmlErrMemory(ctxt, NULL);
3552 if (attvalue != NULL)
3553 xmlFree(attvalue);
3554 goto failed;
3555 }
3556 atts = n;
3557 ctxt->atts = atts;
3558 ctxt->maxatts = maxatts;
3559 }
3560 atts[nbatts++] = attname;
3561 atts[nbatts++] = attvalue;
3562 atts[nbatts] = NULL;
3563 atts[nbatts + 1] = NULL;
3564 }
3565 else {
3566 if (attvalue != NULL)
3567 xmlFree(attvalue);
3568 /* Dump the bogus attribute string up to the next blank or
3569 * the end of the tag. */
3570 while ((IS_CHAR_CH(CUR)) &&
3571 !(IS_BLANK_CH(CUR)) && (CUR != '>') &&
3572 ((CUR != '/') || (NXT(1) != '>')))
3573 NEXT;
3574 }
3575
3576failed:
3577 SKIP_BLANKS;
3578 if (cons == ctxt->nbChars) {
3579 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
3580 "htmlParseStartTag: problem parsing attributes\n",
3581 NULL, NULL);
3582 break;
3583 }
3584 }
3585
3586 /*
3587 * Handle specific association to the META tag
3588 */
3589 if (meta && (nbatts != 0))
3590 htmlCheckMeta(ctxt, atts);
3591
3592 /*
3593 * SAX: Start of Element !
3594 */
3595 htmlnamePush(ctxt, name);
3596 if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL)) {
3597 if (nbatts != 0)
3598 ctxt->sax->startElement(ctxt->userData, name, atts);
3599 else
3600 ctxt->sax->startElement(ctxt->userData, name, NULL);
3601 }
3602
3603 if (atts != NULL) {
3604 for (i = 1;i < nbatts;i += 2) {
3605 if (atts[i] != NULL)
3606 xmlFree((xmlChar *) atts[i]);
3607 }
3608 }
3609
3610 return 0;
3611}
3612
3613/**
3614 * htmlParseEndTag:
3615 * @ctxt: an HTML parser context
3616 *
3617 * parse an end of tag
3618 *
3619 * [42] ETag ::= '</' Name S? '>'
3620 *
3621 * With namespace
3622 *
3623 * [NS 9] ETag ::= '</' QName S? '>'
3624 *
3625 * Returns 1 if the current level should be closed.
3626 */
3627
3628static int
3629htmlParseEndTag(htmlParserCtxtPtr ctxt)
3630{
3631 const xmlChar *name;
3632 const xmlChar *oldname;
3633 int i, ret;
3634
3635 if ((CUR != '<') || (NXT(1) != '/')) {
3636 htmlParseErr(ctxt, XML_ERR_LTSLASH_REQUIRED,
3637 "htmlParseEndTag: '</' not found\n", NULL, NULL);
3638 return (0);
3639 }
3640 SKIP(2);
3641
3642 name = htmlParseHTMLName(ctxt);
3643 if (name == NULL)
3644 return (0);
3645
3646 /*
3647 * We should definitely be at the ending "S? '>'" part
3648 */
3649 SKIP_BLANKS;
3650 if ((!IS_CHAR_CH(CUR)) || (CUR != '>')) {
3651 htmlParseErr(ctxt, XML_ERR_GT_REQUIRED,
3652 "End tag : expected '>'\n", NULL, NULL);
3653 if (ctxt->recovery) {
3654 /*
3655 * We're not at the ending > !!
3656 * Error, unless in recover mode where we search forwards
3657 * until we find a >
3658 */
3659 while (CUR != '\0' && CUR != '>') NEXT;
3660 NEXT;
3661 }
3662 } else
3663 NEXT;
3664
3665 /*
3666 * If the name read is not one of the element in the parsing stack
3667 * then return, it's just an error.
3668 */
3669 for (i = (ctxt->nameNr - 1); i >= 0; i--) {
3670 if (xmlStrEqual(name, ctxt->nameTab[i]))
3671 break;
3672 }
3673 if (i < 0) {
3674 htmlParseErr(ctxt, XML_ERR_TAG_NAME_MISMATCH,
3675 "Unexpected end tag : %s\n", name, NULL);
3676 return (0);
3677 }
3678
3679
3680 /*
3681 * Check for auto-closure of HTML elements.
3682 */
3683
3684 htmlAutoCloseOnClose(ctxt, name);
3685
3686 /*
3687 * Well formedness constraints, opening and closing must match.
3688 * With the exception that the autoclose may have popped stuff out
3689 * of the stack.
3690 */
3691 if (!xmlStrEqual(name, ctxt->name)) {
3692 if ((ctxt->name != NULL) && (!xmlStrEqual(ctxt->name, name))) {
3693 htmlParseErr(ctxt, XML_ERR_TAG_NAME_MISMATCH,
3694 "Opening and ending tag mismatch: %s and %s\n",
3695 name, ctxt->name);
3696 }
3697 }
3698
3699 /*
3700 * SAX: End of Tag
3701 */
3702 oldname = ctxt->name;
3703 if ((oldname != NULL) && (xmlStrEqual(oldname, name))) {
3704 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
3705 ctxt->sax->endElement(ctxt->userData, name);
3706 htmlnamePop(ctxt);
3707 ret = 1;
3708 } else {
3709 ret = 0;
3710 }
3711
3712 return (ret);
3713}
3714
3715
3716/**
3717 * htmlParseReference:
3718 * @ctxt: an HTML parser context
3719 *
3720 * parse and handle entity references in content,
3721 * this will end-up in a call to character() since this is either a
3722 * CharRef, or a predefined entity.
3723 */
3724static void
3725htmlParseReference(htmlParserCtxtPtr ctxt) {
3726 const htmlEntityDesc * ent;
3727 xmlChar out[6];
3728 const xmlChar *name;
3729 if (CUR != '&') return;
3730
3731 if (NXT(1) == '#') {
3732 unsigned int c;
3733 int bits, i = 0;
3734
3735 c = htmlParseCharRef(ctxt);
3736 if (c == 0)
3737 return;
3738
3739 if (c < 0x80) { out[i++]= c; bits= -6; }
3740 else if (c < 0x800) { out[i++]=((c >> 6) & 0x1F) | 0xC0; bits= 0; }
3741 else if (c < 0x10000) { out[i++]=((c >> 12) & 0x0F) | 0xE0; bits= 6; }
3742 else { out[i++]=((c >> 18) & 0x07) | 0xF0; bits= 12; }
3743
3744 for ( ; bits >= 0; bits-= 6) {
3745 out[i++]= ((c >> bits) & 0x3F) | 0x80;
3746 }
3747 out[i] = 0;
3748
3749 htmlCheckParagraph(ctxt);
3750 if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
3751 ctxt->sax->characters(ctxt->userData, out, i);
3752 } else {
3753 ent = htmlParseEntityRef(ctxt, &name);
3754 if (name == NULL) {
3755 htmlCheckParagraph(ctxt);
3756 if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
3757 ctxt->sax->characters(ctxt->userData, BAD_CAST "&", 1);
3758 return;
3759 }
3760 if ((ent == NULL) || !(ent->value > 0)) {
3761 htmlCheckParagraph(ctxt);
3762 if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL)) {
3763 ctxt->sax->characters(ctxt->userData, BAD_CAST "&", 1);
3764 ctxt->sax->characters(ctxt->userData, name, xmlStrlen(name));
3765 /* ctxt->sax->characters(ctxt->userData, BAD_CAST ";", 1); */
3766 }
3767 } else {
3768 unsigned int c;
3769 int bits, i = 0;
3770
3771 c = ent->value;
3772 if (c < 0x80)
3773 { out[i++]= c; bits= -6; }
3774 else if (c < 0x800)
3775 { out[i++]=((c >> 6) & 0x1F) | 0xC0; bits= 0; }
3776 else if (c < 0x10000)
3777 { out[i++]=((c >> 12) & 0x0F) | 0xE0; bits= 6; }
3778 else
3779 { out[i++]=((c >> 18) & 0x07) | 0xF0; bits= 12; }
3780
3781 for ( ; bits >= 0; bits-= 6) {
3782 out[i++]= ((c >> bits) & 0x3F) | 0x80;
3783 }
3784 out[i] = 0;
3785
3786 htmlCheckParagraph(ctxt);
3787 if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
3788 ctxt->sax->characters(ctxt->userData, out, i);
3789 }
3790 }
3791}
3792
3793/**
3794 * htmlParseContent:
3795 * @ctxt: an HTML parser context
3796 *
3797 * Parse a content: comment, sub-element, reference or text.
3798 */
3799
3800static void
3801htmlParseContent(htmlParserCtxtPtr ctxt) {
3802 xmlChar *currentNode;
3803 int depth;
3804 const xmlChar *name;
3805
3806 currentNode = xmlStrdup(ctxt->name);
3807 depth = ctxt->nameNr;
3808 while (1) {
3809 long cons = ctxt->nbChars;
3810
3811 GROW;
3812 /*
3813 * Our tag or one of it's parent or children is ending.
3814 */
3815 if ((CUR == '<') && (NXT(1) == '/')) {
3816 if (htmlParseEndTag(ctxt) &&
3817 ((currentNode != NULL) || (ctxt->nameNr == 0))) {
3818 if (currentNode != NULL)
3819 xmlFree(currentNode);
3820 return;
3821 }
3822 continue; /* while */
3823 }
3824
3825 else if ((CUR == '<') &&
3826 ((IS_ASCII_LETTER(NXT(1))) ||
3827 (NXT(1) == '_') || (NXT(1) == ':'))) {
3828 name = htmlParseHTMLName_nonInvasive(ctxt);
3829 if (name == NULL) {
3830 htmlParseErr(ctxt, XML_ERR_NAME_REQUIRED,
3831 "htmlParseStartTag: invalid element name\n",
3832 NULL, NULL);
3833 /* Dump the bogus tag like browsers do */
3834 while ((IS_CHAR_CH(CUR)) && (CUR != '>'))
3835 NEXT;
3836
3837 if (currentNode != NULL)
3838 xmlFree(currentNode);
3839 return;
3840 }
3841
3842 if (ctxt->name != NULL) {
3843 if (htmlCheckAutoClose(name, ctxt->name) == 1) {
3844 htmlAutoClose(ctxt, name);
3845 continue;
3846 }
3847 }
3848 }
3849
3850 /*
3851 * Has this node been popped out during parsing of
3852 * the next element
3853 */
3854 if ((ctxt->nameNr > 0) && (depth >= ctxt->nameNr) &&
3855 (!xmlStrEqual(currentNode, ctxt->name)))
3856 {
3857 if (currentNode != NULL) xmlFree(currentNode);
3858 return;
3859 }
3860
3861 if ((CUR != 0) && ((xmlStrEqual(currentNode, BAD_CAST"script")) ||
3862 (xmlStrEqual(currentNode, BAD_CAST"style")))) {
3863 /*
3864 * Handle SCRIPT/STYLE separately
3865 */
3866 htmlParseScript(ctxt);
3867 } else {
3868 /*
3869 * Sometimes DOCTYPE arrives in the middle of the document
3870 */
3871 if ((CUR == '<') && (NXT(1) == '!') &&
3872 (UPP(2) == 'D') && (UPP(3) == 'O') &&
3873 (UPP(4) == 'C') && (UPP(5) == 'T') &&
3874 (UPP(6) == 'Y') && (UPP(7) == 'P') &&
3875 (UPP(8) == 'E')) {
3876 htmlParseErr(ctxt, XML_HTML_STRUCURE_ERROR,
3877 "Misplaced DOCTYPE declaration\n",
3878 BAD_CAST "DOCTYPE" , NULL);
3879 htmlParseDocTypeDecl(ctxt);
3880 }
3881
3882 /*
3883 * First case : a comment
3884 */
3885 if ((CUR == '<') && (NXT(1) == '!') &&
3886 (NXT(2) == '-') && (NXT(3) == '-')) {
3887 htmlParseComment(ctxt);
3888 }
3889
3890 /*
3891 * Second case : a Processing Instruction.
3892 */
3893 else if ((CUR == '<') && (NXT(1) == '?')) {
3894 htmlParsePI(ctxt);
3895 }
3896
3897 /*
3898 * Third case : a sub-element.
3899 */
3900 else if (CUR == '<') {
3901 htmlParseElement(ctxt);
3902 }
3903
3904 /*
3905 * Fourth case : a reference. If if has not been resolved,
3906 * parsing returns it's Name, create the node
3907 */
3908 else if (CUR == '&') {
3909 htmlParseReference(ctxt);
3910 }
3911
3912 /*
3913 * Fifth case : end of the resource
3914 */
3915 else if (CUR == 0) {
3916 htmlAutoCloseOnEnd(ctxt);
3917 break;
3918 }
3919
3920 /*
3921 * Last case, text. Note that References are handled directly.
3922 */
3923 else {
3924 htmlParseCharData(ctxt);
3925 }
3926
3927 if (cons == ctxt->nbChars) {
3928 if (ctxt->node != NULL) {
3929 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
3930 "detected an error in element content\n",
3931 NULL, NULL);
3932 }
3933 break;
3934 }
3935 }
3936 GROW;
3937 }
3938 if (currentNode != NULL) xmlFree(currentNode);
3939}
3940
3941/**
3942 * htmlParseContent:
3943 * @ctxt: an HTML parser context
3944 *
3945 * Parse a content: comment, sub-element, reference or text.
3946 */
3947
3948void
3949__htmlParseContent(void *ctxt) {
3950 if (ctxt != NULL)
3951 htmlParseContent((htmlParserCtxtPtr) ctxt);
3952}
3953
3954/**
3955 * htmlParseElement:
3956 * @ctxt: an HTML parser context
3957 *
3958 * parse an HTML element, this is highly recursive
3959 *
3960 * [39] element ::= EmptyElemTag | STag content ETag
3961 *
3962 * [41] Attribute ::= Name Eq AttValue
3963 */
3964
3965void
3966htmlParseElement(htmlParserCtxtPtr ctxt) {
3967 const xmlChar *name;
3968 xmlChar *currentNode = NULL;
3969 const htmlElemDesc * info;
3970 htmlParserNodeInfo node_info;
3971 int failed;
3972 int depth;
3973 const xmlChar *oldptr;
3974
3975 if ((ctxt == NULL) || (ctxt->input == NULL)) {
3976 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
3977 "htmlParseElement: context error\n", NULL, NULL);
3978 return;
3979 }
3980 /* Capture start position */
3981 if (ctxt->record_info) {
3982 node_info.begin_pos = ctxt->input->consumed +
3983 (CUR_PTR - ctxt->input->base);
3984 node_info.begin_line = ctxt->input->line;
3985 }
3986
3987 failed = htmlParseStartTag(ctxt);
3988 name = ctxt->name;
3989 if (failed || (name == NULL)) {
3990 if (CUR == '>')
3991 NEXT;
3992 return;
3993 }
3994
3995 /*
3996 * Lookup the info for that element.
3997 */
3998 info = htmlTagLookup(name);
3999 if (info == NULL) {
4000 htmlParseErr(ctxt, XML_HTML_UNKNOWN_TAG,
4001 "Tag %s invalid\n", name, NULL);
4002 }
4003
4004 /*
4005 * Check for an Empty Element labeled the XML/SGML way
4006 */
4007 if ((CUR == '/') && (NXT(1) == '>')) {
4008 SKIP(2);
4009 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
4010 ctxt->sax->endElement(ctxt->userData, name);
4011 htmlnamePop(ctxt);
4012 return;
4013 }
4014
4015 if (CUR == '>') {
4016 NEXT;
4017 } else {
4018 htmlParseErr(ctxt, XML_ERR_GT_REQUIRED,
4019 "Couldn't find end of Start Tag %s\n", name, NULL);
4020
4021 /*
4022 * end of parsing of this node.
4023 */
4024 if (xmlStrEqual(name, ctxt->name)) {
4025 nodePop(ctxt);
4026 htmlnamePop(ctxt);
4027 }
4028
4029 /*
4030 * Capture end position and add node
4031 */
4032 if (ctxt->record_info) {
4033 node_info.end_pos = ctxt->input->consumed +
4034 (CUR_PTR - ctxt->input->base);
4035 node_info.end_line = ctxt->input->line;
4036 node_info.node = ctxt->node;
4037 xmlParserAddNodeInfo(ctxt, &node_info);
4038 }
4039 return;
4040 }
4041
4042 /*
4043 * Check for an Empty Element from DTD definition
4044 */
4045 if ((info != NULL) && (info->empty)) {
4046 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
4047 ctxt->sax->endElement(ctxt->userData, name);
4048 htmlnamePop(ctxt);
4049 return;
4050 }
4051
4052 /*
4053 * Parse the content of the element:
4054 */
4055 currentNode = xmlStrdup(ctxt->name);
4056 depth = ctxt->nameNr;
4057 while (IS_CHAR_CH(CUR)) {
4058 oldptr = ctxt->input->cur;
4059 htmlParseContent(ctxt);
4060 if (oldptr==ctxt->input->cur) break;
4061 if (ctxt->nameNr < depth) break;
4062 }
4063
4064 /*
4065 * Capture end position and add node
4066 */
4067 if ( currentNode != NULL && ctxt->record_info ) {
4068 node_info.end_pos = ctxt->input->consumed +
4069 (CUR_PTR - ctxt->input->base);
4070 node_info.end_line = ctxt->input->line;
4071 node_info.node = ctxt->node;
4072 xmlParserAddNodeInfo(ctxt, &node_info);
4073 }
4074 if (!IS_CHAR_CH(CUR)) {
4075 htmlAutoCloseOnEnd(ctxt);
4076 }
4077
4078 if (currentNode != NULL)
4079 xmlFree(currentNode);
4080}
4081
4082/**
4083 * htmlParseDocument:
4084 * @ctxt: an HTML parser context
4085 *
4086 * parse an HTML document (and build a tree if using the standard SAX
4087 * interface).
4088 *
4089 * Returns 0, -1 in case of error. the parser context is augmented
4090 * as a result of the parsing.
4091 */
4092
4093int
4094htmlParseDocument(htmlParserCtxtPtr ctxt) {
4095 xmlDtdPtr dtd;
4096
4097 xmlInitParser();
4098
4099 htmlDefaultSAXHandlerInit();
4100
4101 if ((ctxt == NULL) || (ctxt->input == NULL)) {
4102 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
4103 "htmlParseDocument: context error\n", NULL, NULL);
4104 return(XML_ERR_INTERNAL_ERROR);
4105 }
4106 ctxt->html = 1;
4107 GROW;
4108 /*
4109 * SAX: beginning of the document processing.
4110 */
4111 if ((ctxt->sax) && (ctxt->sax->setDocumentLocator))
4112 ctxt->sax->setDocumentLocator(ctxt->userData, &xmlDefaultSAXLocator);
4113
4114 /*
4115 * Wipe out everything which is before the first '<'
4116 */
4117 SKIP_BLANKS;
4118 if (CUR == 0) {
4119 htmlParseErr(ctxt, XML_ERR_DOCUMENT_EMPTY,
4120 "Document is empty\n", NULL, NULL);
4121 }
4122
4123 if ((ctxt->sax) && (ctxt->sax->startDocument) && (!ctxt->disableSAX))
4124 ctxt->sax->startDocument(ctxt->userData);
4125
4126
4127 /*
4128 * Parse possible comments and PIs before any content
4129 */
4130 while (((CUR == '<') && (NXT(1) == '!') &&
4131 (NXT(2) == '-') && (NXT(3) == '-')) ||
4132 ((CUR == '<') && (NXT(1) == '?'))) {
4133 htmlParseComment(ctxt);
4134 htmlParsePI(ctxt);
4135 SKIP_BLANKS;
4136 }
4137
4138
4139 /*
4140 * Then possibly doc type declaration(s) and more Misc
4141 * (doctypedecl Misc*)?
4142 */
4143 if ((CUR == '<') && (NXT(1) == '!') &&
4144 (UPP(2) == 'D') && (UPP(3) == 'O') &&
4145 (UPP(4) == 'C') && (UPP(5) == 'T') &&
4146 (UPP(6) == 'Y') && (UPP(7) == 'P') &&
4147 (UPP(8) == 'E')) {
4148 htmlParseDocTypeDecl(ctxt);
4149 }
4150 SKIP_BLANKS;
4151
4152 /*
4153 * Parse possible comments and PIs before any content
4154 */
4155 while (((CUR == '<') && (NXT(1) == '!') &&
4156 (NXT(2) == '-') && (NXT(3) == '-')) ||
4157 ((CUR == '<') && (NXT(1) == '?'))) {
4158 htmlParseComment(ctxt);
4159 htmlParsePI(ctxt);
4160 SKIP_BLANKS;
4161 }
4162
4163 /*
4164 * Time to start parsing the tree itself
4165 */
4166 htmlParseContent(ctxt);
4167
4168 /*
4169 * autoclose
4170 */
4171 if (CUR == 0)
4172 htmlAutoCloseOnEnd(ctxt);
4173
4174
4175 /*
4176 * SAX: end of the document processing.
4177 */
4178 if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
4179 ctxt->sax->endDocument(ctxt->userData);
4180
4181 if (ctxt->myDoc != NULL) {
4182 dtd = xmlGetIntSubset(ctxt->myDoc);
4183 if (dtd == NULL)
4184 ctxt->myDoc->intSubset =
4185 xmlCreateIntSubset(ctxt->myDoc, BAD_CAST "html",
4186 BAD_CAST "-//W3C//DTD HTML 4.0 Transitional//EN",
4187 BAD_CAST "http://www.w3.org/TR/REC-html40/loose.dtd");
4188 }
4189 if (! ctxt->wellFormed) return(-1);
4190 return(0);
4191}
4192
4193
4194/************************************************************************
4195 * *
4196 * Parser contexts handling *
4197 * *
4198 ************************************************************************/
4199
4200/**
4201 * htmlInitParserCtxt:
4202 * @ctxt: an HTML parser context
4203 *
4204 * Initialize a parser context
4205 *
4206 * Returns 0 in case of success and -1 in case of error
4207 */
4208
4209static int
4210htmlInitParserCtxt(htmlParserCtxtPtr ctxt)
4211{
4212 htmlSAXHandler *sax;
4213
4214 if (ctxt == NULL) return(-1);
4215 memset(ctxt, 0, sizeof(htmlParserCtxt));
4216
4217 ctxt->dict = xmlDictCreate();
4218 if (ctxt->dict == NULL) {
4219 htmlErrMemory(NULL, "htmlInitParserCtxt: out of memory\n");
4220 return(-1);
4221 }
4222 sax = (htmlSAXHandler *) xmlMalloc(sizeof(htmlSAXHandler));
4223 if (sax == NULL) {
4224 htmlErrMemory(NULL, "htmlInitParserCtxt: out of memory\n");
4225 return(-1);
4226 }
4227 else
4228 memset(sax, 0, sizeof(htmlSAXHandler));
4229
4230 /* Allocate the Input stack */
4231 ctxt->inputTab = (htmlParserInputPtr *)
4232 xmlMalloc(5 * sizeof(htmlParserInputPtr));
4233 if (ctxt->inputTab == NULL) {
4234 htmlErrMemory(NULL, "htmlInitParserCtxt: out of memory\n");
4235 ctxt->inputNr = 0;
4236 ctxt->inputMax = 0;
4237 ctxt->input = NULL;
4238 return(-1);
4239 }
4240 ctxt->inputNr = 0;
4241 ctxt->inputMax = 5;
4242 ctxt->input = NULL;
4243 ctxt->version = NULL;
4244 ctxt->encoding = NULL;
4245 ctxt->standalone = -1;
4246 ctxt->instate = XML_PARSER_START;
4247
4248 /* Allocate the Node stack */
4249 ctxt->nodeTab = (htmlNodePtr *) xmlMalloc(10 * sizeof(htmlNodePtr));
4250 if (ctxt->nodeTab == NULL) {
4251 htmlErrMemory(NULL, "htmlInitParserCtxt: out of memory\n");
4252 ctxt->nodeNr = 0;
4253 ctxt->nodeMax = 0;
4254 ctxt->node = NULL;
4255 ctxt->inputNr = 0;
4256 ctxt->inputMax = 0;
4257 ctxt->input = NULL;
4258 return(-1);
4259 }
4260 ctxt->nodeNr = 0;
4261 ctxt->nodeMax = 10;
4262 ctxt->node = NULL;
4263
4264 /* Allocate the Name stack */
4265 ctxt->nameTab = (const xmlChar **) xmlMalloc(10 * sizeof(xmlChar *));
4266 if (ctxt->nameTab == NULL) {
4267 htmlErrMemory(NULL, "htmlInitParserCtxt: out of memory\n");
4268 ctxt->nameNr = 0;
4269 ctxt->nameMax = 10;
4270 ctxt->name = NULL;
4271 ctxt->nodeNr = 0;
4272 ctxt->nodeMax = 0;
4273 ctxt->node = NULL;
4274 ctxt->inputNr = 0;
4275 ctxt->inputMax = 0;
4276 ctxt->input = NULL;
4277 return(-1);
4278 }
4279 ctxt->nameNr = 0;
4280 ctxt->nameMax = 10;
4281 ctxt->name = NULL;
4282
4283 if (sax == NULL) ctxt->sax = (xmlSAXHandlerPtr) &htmlDefaultSAXHandler;
4284 else {
4285 ctxt->sax = sax;
4286 memcpy(sax, &htmlDefaultSAXHandler, sizeof(xmlSAXHandlerV1));
4287 }
4288 ctxt->userData = ctxt;
4289 ctxt->myDoc = NULL;
4290 ctxt->wellFormed = 1;
4291 ctxt->replaceEntities = 0;
4292 ctxt->linenumbers = xmlLineNumbersDefaultValue;
4293 ctxt->html = 1;
4294 ctxt->vctxt.finishDtd = XML_CTXT_FINISH_DTD_0;
4295 ctxt->vctxt.userData = ctxt;
4296 ctxt->vctxt.error = xmlParserValidityError;
4297 ctxt->vctxt.warning = xmlParserValidityWarning;
4298 ctxt->record_info = 0;
4299 ctxt->validate = 0;
4300 ctxt->nbChars = 0;
4301 ctxt->checkIndex = 0;
4302 ctxt->catalogs = NULL;
4303 xmlInitNodeInfoSeq(&ctxt->node_seq);
4304 return(0);
4305}
4306
4307/**
4308 * htmlFreeParserCtxt:
4309 * @ctxt: an HTML parser context
4310 *
4311 * Free all the memory used by a parser context. However the parsed
4312 * document in ctxt->myDoc is not freed.
4313 */
4314
4315void
4316htmlFreeParserCtxt(htmlParserCtxtPtr ctxt)
4317{
4318 xmlFreeParserCtxt(ctxt);
4319}
4320
4321/**
4322 * htmlNewParserCtxt:
4323 *
4324 * Allocate and initialize a new parser context.
4325 *
4326 * Returns the htmlParserCtxtPtr or NULL in case of allocation error
4327 */
4328
4329htmlParserCtxtPtr
4330htmlNewParserCtxt(void)
4331{
4332 xmlParserCtxtPtr ctxt;
4333
4334 ctxt = (xmlParserCtxtPtr) xmlMalloc(sizeof(xmlParserCtxt));
4335 if (ctxt == NULL) {
4336 htmlErrMemory(NULL, "NewParserCtxt: out of memory\n");
4337 return(NULL);
4338 }
4339 memset(ctxt, 0, sizeof(xmlParserCtxt));
4340 if (htmlInitParserCtxt(ctxt) < 0) {
4341 htmlFreeParserCtxt(ctxt);
4342 return(NULL);
4343 }
4344 return(ctxt);
4345}
4346
4347/**
4348 * htmlCreateMemoryParserCtxt:
4349 * @buffer: a pointer to a char array
4350 * @size: the size of the array
4351 *
4352 * Create a parser context for an HTML in-memory document.
4353 *
4354 * Returns the new parser context or NULL
4355 */
4356htmlParserCtxtPtr
4357htmlCreateMemoryParserCtxt(const char *buffer, int size) {
4358 xmlParserCtxtPtr ctxt;
4359 xmlParserInputPtr input;
4360 xmlParserInputBufferPtr buf;
4361
4362 if (buffer == NULL)
4363 return(NULL);
4364 if (size <= 0)
4365 return(NULL);
4366
4367 ctxt = htmlNewParserCtxt();
4368 if (ctxt == NULL)
4369 return(NULL);
4370
4371 buf = xmlParserInputBufferCreateMem(buffer, size, XML_CHAR_ENCODING_NONE);
4372 if (buf == NULL) return(NULL);
4373
4374 input = xmlNewInputStream(ctxt);
4375 if (input == NULL) {
4376 xmlFreeParserCtxt(ctxt);
4377 return(NULL);
4378 }
4379
4380 input->filename = NULL;
4381 input->buf = buf;
4382 input->base = input->buf->buffer->content;
4383 input->cur = input->buf->buffer->content;
4384 input->end = &input->buf->buffer->content[input->buf->buffer->use];
4385
4386 inputPush(ctxt, input);
4387 return(ctxt);
4388}
4389
4390/**
4391 * htmlCreateDocParserCtxt:
4392 * @cur: a pointer to an array of xmlChar
4393 * @encoding: a free form C string describing the HTML document encoding, or NULL
4394 *
4395 * Create a parser context for an HTML document.
4396 *
4397 * TODO: check the need to add encoding handling there
4398 *
4399 * Returns the new parser context or NULL
4400 */
4401static htmlParserCtxtPtr
4402htmlCreateDocParserCtxt(const xmlChar *cur, const char *encoding) {
4403 int len;
4404 htmlParserCtxtPtr ctxt;
4405
4406 if (cur == NULL)
4407 return(NULL);
4408 len = xmlStrlen(cur);
4409 ctxt = htmlCreateMemoryParserCtxt((char *)cur, len);
4410 if (ctxt == NULL)
4411 return(NULL);
4412
4413 if (encoding != NULL) {
4414 xmlCharEncoding enc;
4415 xmlCharEncodingHandlerPtr handler;
4416
4417 if (ctxt->input->encoding != NULL)
4418 xmlFree((xmlChar *) ctxt->input->encoding);
4419 ctxt->input->encoding = xmlStrdup((const xmlChar *) encoding);
4420
4421 enc = xmlParseCharEncoding(encoding);
4422 /*
4423 * registered set of known encodings
4424 */
4425 if (enc != XML_CHAR_ENCODING_ERROR) {
4426 xmlSwitchEncoding(ctxt, enc);
4427 if (ctxt->errNo == XML_ERR_UNSUPPORTED_ENCODING) {
4428 htmlParseErr(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
4429 "Unsupported encoding %s\n",
4430 (const xmlChar *) encoding, NULL);
4431 }
4432 } else {
4433 /*
4434 * fallback for unknown encodings
4435 */
4436 handler = xmlFindCharEncodingHandler((const char *) encoding);
4437 if (handler != NULL) {
4438 xmlSwitchToEncoding(ctxt, handler);
4439 } else {
4440 htmlParseErr(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
4441 "Unsupported encoding %s\n",
4442 (const xmlChar *) encoding, NULL);
4443 }
4444 }
4445 }
4446 return(ctxt);
4447}
4448
4449#ifdef LIBXML_PUSH_ENABLED
4450/************************************************************************
4451 * *
4452 * Progressive parsing interfaces *
4453 * *
4454 ************************************************************************/
4455
4456/**
4457 * htmlParseLookupSequence:
4458 * @ctxt: an HTML parser context
4459 * @first: the first char to lookup
4460 * @next: the next char to lookup or zero
4461 * @third: the next char to lookup or zero
4462 * @comment: flag to force checking inside comments
4463 *
4464 * Try to find if a sequence (first, next, third) or just (first next) or
4465 * (first) is available in the input stream.
4466 * This function has a side effect of (possibly) incrementing ctxt->checkIndex
4467 * to avoid rescanning sequences of bytes, it DOES change the state of the
4468 * parser, do not use liberally.
4469 * This is basically similar to xmlParseLookupSequence()
4470 *
4471 * Returns the index to the current parsing point if the full sequence
4472 * is available, -1 otherwise.
4473 */
4474static int
4475htmlParseLookupSequence(htmlParserCtxtPtr ctxt, xmlChar first,
4476 xmlChar next, xmlChar third, int iscomment) {
4477 int base, len;
4478 htmlParserInputPtr in;
4479 const xmlChar *buf;
4480 int incomment = 0;
4481
4482 in = ctxt->input;
4483 if (in == NULL) return(-1);
4484 base = in->cur - in->base;
4485 if (base < 0) return(-1);
4486 if (ctxt->checkIndex > base)
4487 base = ctxt->checkIndex;
4488 if (in->buf == NULL) {
4489 buf = in->base;
4490 len = in->length;
4491 } else {
4492 buf = in->buf->buffer->content;
4493 len = in->buf->buffer->use;
4494 }
4495 /* take into account the sequence length */
4496 if (third) len -= 2;
4497 else if (next) len --;
4498 for (;base < len;base++) {
4499 if (!incomment && (base + 4 < len) && !iscomment) {
4500 if ((buf[base] == '<') && (buf[base + 1] == '!') &&
4501 (buf[base + 2] == '-') && (buf[base + 3] == '-')) {
4502 incomment = 1;
4503 /* do not increment past <! - some people use <!--> */
4504 base += 2;
4505 }
4506 }
4507 if (incomment) {
4508 if (base + 3 > len)
4509 return(-1);
4510 if ((buf[base] == '-') && (buf[base + 1] == '-') &&
4511 (buf[base + 2] == '>')) {
4512 incomment = 0;
4513 base += 2;
4514 }
4515 continue;
4516 }
4517 if (buf[base] == first) {
4518 if (third != 0) {
4519 if ((buf[base + 1] != next) ||
4520 (buf[base + 2] != third)) continue;
4521 } else if (next != 0) {
4522 if (buf[base + 1] != next) continue;
4523 }
4524 ctxt->checkIndex = 0;
4525#ifdef DEBUG_PUSH
4526 if (next == 0)
4527 xmlGenericError(xmlGenericErrorContext,
4528 "HPP: lookup '%c' found at %d\n",
4529 first, base);
4530 else if (third == 0)
4531 xmlGenericError(xmlGenericErrorContext,
4532 "HPP: lookup '%c%c' found at %d\n",
4533 first, next, base);
4534 else
4535 xmlGenericError(xmlGenericErrorContext,
4536 "HPP: lookup '%c%c%c' found at %d\n",
4537 first, next, third, base);
4538#endif
4539 return(base - (in->cur - in->base));
4540 }
4541 }
4542 ctxt->checkIndex = base;
4543#ifdef DEBUG_PUSH
4544 if (next == 0)
4545 xmlGenericError(xmlGenericErrorContext,
4546 "HPP: lookup '%c' failed\n", first);
4547 else if (third == 0)
4548 xmlGenericError(xmlGenericErrorContext,
4549 "HPP: lookup '%c%c' failed\n", first, next);
4550 else
4551 xmlGenericError(xmlGenericErrorContext,
4552 "HPP: lookup '%c%c%c' failed\n", first, next, third);
4553#endif
4554 return(-1);
4555}
4556
4557/**
4558 * htmlParseTryOrFinish:
4559 * @ctxt: an HTML parser context
4560 * @terminate: last chunk indicator
4561 *
4562 * Try to progress on parsing
4563 *
4564 * Returns zero if no parsing was possible
4565 */
4566static int
4567htmlParseTryOrFinish(htmlParserCtxtPtr ctxt, int terminate) {
4568 int ret = 0;
4569 htmlParserInputPtr in;
4570 int avail = 0;
4571 xmlChar cur, next;
4572
4573#ifdef DEBUG_PUSH
4574 switch (ctxt->instate) {
4575 case XML_PARSER_EOF:
4576 xmlGenericError(xmlGenericErrorContext,
4577 "HPP: try EOF\n"); break;
4578 case XML_PARSER_START:
4579 xmlGenericError(xmlGenericErrorContext,
4580 "HPP: try START\n"); break;
4581 case XML_PARSER_MISC:
4582 xmlGenericError(xmlGenericErrorContext,
4583 "HPP: try MISC\n");break;
4584 case XML_PARSER_COMMENT:
4585 xmlGenericError(xmlGenericErrorContext,
4586 "HPP: try COMMENT\n");break;
4587 case XML_PARSER_PROLOG:
4588 xmlGenericError(xmlGenericErrorContext,
4589 "HPP: try PROLOG\n");break;
4590 case XML_PARSER_START_TAG:
4591 xmlGenericError(xmlGenericErrorContext,
4592 "HPP: try START_TAG\n");break;
4593 case XML_PARSER_CONTENT:
4594 xmlGenericError(xmlGenericErrorContext,
4595 "HPP: try CONTENT\n");break;
4596 case XML_PARSER_CDATA_SECTION:
4597 xmlGenericError(xmlGenericErrorContext,
4598 "HPP: try CDATA_SECTION\n");break;
4599 case XML_PARSER_END_TAG:
4600 xmlGenericError(xmlGenericErrorContext,
4601 "HPP: try END_TAG\n");break;
4602 case XML_PARSER_ENTITY_DECL:
4603 xmlGenericError(xmlGenericErrorContext,
4604 "HPP: try ENTITY_DECL\n");break;
4605 case XML_PARSER_ENTITY_VALUE:
4606 xmlGenericError(xmlGenericErrorContext,
4607 "HPP: try ENTITY_VALUE\n");break;
4608 case XML_PARSER_ATTRIBUTE_VALUE:
4609 xmlGenericError(xmlGenericErrorContext,
4610 "HPP: try ATTRIBUTE_VALUE\n");break;
4611 case XML_PARSER_DTD:
4612 xmlGenericError(xmlGenericErrorContext,
4613 "HPP: try DTD\n");break;
4614 case XML_PARSER_EPILOG:
4615 xmlGenericError(xmlGenericErrorContext,
4616 "HPP: try EPILOG\n");break;
4617 case XML_PARSER_PI:
4618 xmlGenericError(xmlGenericErrorContext,
4619 "HPP: try PI\n");break;
4620 case XML_PARSER_SYSTEM_LITERAL:
4621 xmlGenericError(xmlGenericErrorContext,
4622 "HPP: try SYSTEM_LITERAL\n");break;
4623 }
4624#endif
4625
4626 while (1) {
4627
4628 in = ctxt->input;
4629 if (in == NULL) break;
4630 if (in->buf == NULL)
4631 avail = in->length - (in->cur - in->base);
4632 else
4633 avail = in->buf->buffer->use - (in->cur - in->base);
4634 if ((avail == 0) && (terminate)) {
4635 htmlAutoCloseOnEnd(ctxt);
4636 if ((ctxt->nameNr == 0) && (ctxt->instate != XML_PARSER_EOF)) {
4637 /*
4638 * SAX: end of the document processing.
4639 */
4640 ctxt->instate = XML_PARSER_EOF;
4641 if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
4642 ctxt->sax->endDocument(ctxt->userData);
4643 }
4644 }
4645 if (avail < 1)
4646 goto done;
4647 cur = in->cur[0];
4648 if (cur == 0) {
4649 SKIP(1);
4650 continue;
4651 }
4652
4653 switch (ctxt->instate) {
4654 case XML_PARSER_EOF:
4655 /*
4656 * Document parsing is done !
4657 */
4658 goto done;
4659 case XML_PARSER_START:
4660 /*
4661 * Very first chars read from the document flow.
4662 */
4663 cur = in->cur[0];
4664 if (IS_BLANK_CH(cur)) {
4665 SKIP_BLANKS;
4666 if (in->buf == NULL)
4667 avail = in->length - (in->cur - in->base);
4668 else
4669 avail = in->buf->buffer->use - (in->cur - in->base);
4670 }
4671 if ((ctxt->sax) && (ctxt->sax->setDocumentLocator))
4672 ctxt->sax->setDocumentLocator(ctxt->userData,
4673 &xmlDefaultSAXLocator);
4674 if ((ctxt->sax) && (ctxt->sax->startDocument) &&
4675 (!ctxt->disableSAX))
4676 ctxt->sax->startDocument(ctxt->userData);
4677
4678 cur = in->cur[0];
4679 next = in->cur[1];
4680 if ((cur == '<') && (next == '!') &&
4681 (UPP(2) == 'D') && (UPP(3) == 'O') &&
4682 (UPP(4) == 'C') && (UPP(5) == 'T') &&
4683 (UPP(6) == 'Y') && (UPP(7) == 'P') &&
4684 (UPP(8) == 'E')) {
4685 if ((!terminate) &&
4686 (htmlParseLookupSequence(ctxt, '>', 0, 0, 0) < 0))
4687 goto done;
4688#ifdef DEBUG_PUSH
4689 xmlGenericError(xmlGenericErrorContext,
4690 "HPP: Parsing internal subset\n");
4691#endif
4692 htmlParseDocTypeDecl(ctxt);
4693 ctxt->instate = XML_PARSER_PROLOG;
4694#ifdef DEBUG_PUSH
4695 xmlGenericError(xmlGenericErrorContext,
4696 "HPP: entering PROLOG\n");
4697#endif
4698 } else {
4699 ctxt->instate = XML_PARSER_MISC;
4700#ifdef DEBUG_PUSH
4701 xmlGenericError(xmlGenericErrorContext,
4702 "HPP: entering MISC\n");
4703#endif
4704 }
4705 break;
4706 case XML_PARSER_MISC:
4707 SKIP_BLANKS;
4708 if (in->buf == NULL)
4709 avail = in->length - (in->cur - in->base);
4710 else
4711 avail = in->buf->buffer->use - (in->cur - in->base);
4712 if (avail < 2)
4713 goto done;
4714 cur = in->cur[0];
4715 next = in->cur[1];
4716 if ((cur == '<') && (next == '!') &&
4717 (in->cur[2] == '-') && (in->cur[3] == '-')) {
4718 if ((!terminate) &&
4719 (htmlParseLookupSequence(ctxt, '-', '-', '>', 1) < 0))
4720 goto done;
4721#ifdef DEBUG_PUSH
4722 xmlGenericError(xmlGenericErrorContext,
4723 "HPP: Parsing Comment\n");
4724#endif
4725 htmlParseComment(ctxt);
4726 ctxt->instate = XML_PARSER_MISC;
4727 } else if ((cur == '<') && (next == '?')) {
4728 if ((!terminate) &&
4729 (htmlParseLookupSequence(ctxt, '>', 0, 0, 0) < 0))
4730 goto done;
4731#ifdef DEBUG_PUSH
4732 xmlGenericError(xmlGenericErrorContext,
4733 "HPP: Parsing PI\n");
4734#endif
4735 htmlParsePI(ctxt);
4736 ctxt->instate = XML_PARSER_MISC;
4737 } else if ((cur == '<') && (next == '!') &&
4738 (UPP(2) == 'D') && (UPP(3) == 'O') &&
4739 (UPP(4) == 'C') && (UPP(5) == 'T') &&
4740 (UPP(6) == 'Y') && (UPP(7) == 'P') &&
4741 (UPP(8) == 'E')) {
4742 if ((!terminate) &&
4743 (htmlParseLookupSequence(ctxt, '>', 0, 0, 0) < 0))
4744 goto done;
4745#ifdef DEBUG_PUSH
4746 xmlGenericError(xmlGenericErrorContext,
4747 "HPP: Parsing internal subset\n");
4748#endif
4749 htmlParseDocTypeDecl(ctxt);
4750 ctxt->instate = XML_PARSER_PROLOG;
4751#ifdef DEBUG_PUSH
4752 xmlGenericError(xmlGenericErrorContext,
4753 "HPP: entering PROLOG\n");
4754#endif
4755 } else if ((cur == '<') && (next == '!') &&
4756 (avail < 9)) {
4757 goto done;
4758 } else {
4759 ctxt->instate = XML_PARSER_START_TAG;
4760#ifdef DEBUG_PUSH
4761 xmlGenericError(xmlGenericErrorContext,
4762 "HPP: entering START_TAG\n");
4763#endif
4764 }
4765 break;
4766 case XML_PARSER_PROLOG:
4767 SKIP_BLANKS;
4768 if (in->buf == NULL)
4769 avail = in->length - (in->cur - in->base);
4770 else
4771 avail = in->buf->buffer->use - (in->cur - in->base);
4772 if (avail < 2)
4773 goto done;
4774 cur = in->cur[0];
4775 next = in->cur[1];
4776 if ((cur == '<') && (next == '!') &&
4777 (in->cur[2] == '-') && (in->cur[3] == '-')) {
4778 if ((!terminate) &&
4779 (htmlParseLookupSequence(ctxt, '-', '-', '>', 1) < 0))
4780 goto done;
4781#ifdef DEBUG_PUSH
4782 xmlGenericError(xmlGenericErrorContext,
4783 "HPP: Parsing Comment\n");
4784#endif
4785 htmlParseComment(ctxt);
4786 ctxt->instate = XML_PARSER_PROLOG;
4787 } else if ((cur == '<') && (next == '?')) {
4788 if ((!terminate) &&
4789 (htmlParseLookupSequence(ctxt, '>', 0, 0, 0) < 0))
4790 goto done;
4791#ifdef DEBUG_PUSH
4792 xmlGenericError(xmlGenericErrorContext,
4793 "HPP: Parsing PI\n");
4794#endif
4795 htmlParsePI(ctxt);
4796 ctxt->instate = XML_PARSER_PROLOG;
4797 } else if ((cur == '<') && (next == '!') &&
4798 (avail < 4)) {
4799 goto done;
4800 } else {
4801 ctxt->instate = XML_PARSER_START_TAG;
4802#ifdef DEBUG_PUSH
4803 xmlGenericError(xmlGenericErrorContext,
4804 "HPP: entering START_TAG\n");
4805#endif
4806 }
4807 break;
4808 case XML_PARSER_EPILOG:
4809 if (in->buf == NULL)
4810 avail = in->length - (in->cur - in->base);
4811 else
4812 avail = in->buf->buffer->use - (in->cur - in->base);
4813 if (avail < 1)
4814 goto done;
4815 cur = in->cur[0];
4816 if (IS_BLANK_CH(cur)) {
4817 htmlParseCharData(ctxt);
4818 goto done;
4819 }
4820 if (avail < 2)
4821 goto done;
4822 next = in->cur[1];
4823 if ((cur == '<') && (next == '!') &&
4824 (in->cur[2] == '-') && (in->cur[3] == '-')) {
4825 if ((!terminate) &&
4826 (htmlParseLookupSequence(ctxt, '-', '-', '>', 1) < 0))
4827 goto done;
4828#ifdef DEBUG_PUSH
4829 xmlGenericError(xmlGenericErrorContext,
4830 "HPP: Parsing Comment\n");
4831#endif
4832 htmlParseComment(ctxt);
4833 ctxt->instate = XML_PARSER_EPILOG;
4834 } else if ((cur == '<') && (next == '?')) {
4835 if ((!terminate) &&
4836 (htmlParseLookupSequence(ctxt, '>', 0, 0, 0) < 0))
4837 goto done;
4838#ifdef DEBUG_PUSH
4839 xmlGenericError(xmlGenericErrorContext,
4840 "HPP: Parsing PI\n");
4841#endif
4842 htmlParsePI(ctxt);
4843 ctxt->instate = XML_PARSER_EPILOG;
4844 } else if ((cur == '<') && (next == '!') &&
4845 (avail < 4)) {
4846 goto done;
4847 } else {
4848 ctxt->errNo = XML_ERR_DOCUMENT_END;
4849 ctxt->wellFormed = 0;
4850 ctxt->instate = XML_PARSER_EOF;
4851#ifdef DEBUG_PUSH
4852 xmlGenericError(xmlGenericErrorContext,
4853 "HPP: entering EOF\n");
4854#endif
4855 if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
4856 ctxt->sax->endDocument(ctxt->userData);
4857 goto done;
4858 }
4859 break;
4860 case XML_PARSER_START_TAG: {
4861 const xmlChar *name;
4862 int failed;
4863 const htmlElemDesc * info;
4864
4865 if (avail < 2)
4866 goto done;
4867 cur = in->cur[0];
4868 if (cur != '<') {
4869 ctxt->instate = XML_PARSER_CONTENT;
4870#ifdef DEBUG_PUSH
4871 xmlGenericError(xmlGenericErrorContext,
4872 "HPP: entering CONTENT\n");
4873#endif
4874 break;
4875 }
4876 if (in->cur[1] == '/') {
4877 ctxt->instate = XML_PARSER_END_TAG;
4878 ctxt->checkIndex = 0;
4879#ifdef DEBUG_PUSH
4880 xmlGenericError(xmlGenericErrorContext,
4881 "HPP: entering END_TAG\n");
4882#endif
4883 break;
4884 }
4885 if ((!terminate) &&
4886 (htmlParseLookupSequence(ctxt, '>', 0, 0, 0) < 0))
4887 goto done;
4888
4889 failed = htmlParseStartTag(ctxt);
4890 name = ctxt->name;
4891 if (failed ||
4892 (name == NULL)) {
4893 if (CUR == '>')
4894 NEXT;
4895 break;
4896 }
4897
4898 /*
4899 * Lookup the info for that element.
4900 */
4901 info = htmlTagLookup(name);
4902 if (info == NULL) {
4903 htmlParseErr(ctxt, XML_HTML_UNKNOWN_TAG,
4904 "Tag %s invalid\n", name, NULL);
4905 }
4906
4907 /*
4908 * Check for an Empty Element labeled the XML/SGML way
4909 */
4910 if ((CUR == '/') && (NXT(1) == '>')) {
4911 SKIP(2);
4912 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
4913 ctxt->sax->endElement(ctxt->userData, name);
4914 htmlnamePop(ctxt);
4915 ctxt->instate = XML_PARSER_CONTENT;
4916#ifdef DEBUG_PUSH
4917 xmlGenericError(xmlGenericErrorContext,
4918 "HPP: entering CONTENT\n");
4919#endif
4920 break;
4921 }
4922
4923 if (CUR == '>') {
4924 NEXT;
4925 } else {
4926 htmlParseErr(ctxt, XML_ERR_GT_REQUIRED,
4927 "Couldn't find end of Start Tag %s\n",
4928 name, NULL);
4929
4930 /*
4931 * end of parsing of this node.
4932 */
4933 if (xmlStrEqual(name, ctxt->name)) {
4934 nodePop(ctxt);
4935 htmlnamePop(ctxt);
4936 }
4937
4938 ctxt->instate = XML_PARSER_CONTENT;
4939#ifdef DEBUG_PUSH
4940 xmlGenericError(xmlGenericErrorContext,
4941 "HPP: entering CONTENT\n");
4942#endif
4943 break;
4944 }
4945
4946 /*
4947 * Check for an Empty Element from DTD definition
4948 */
4949 if ((info != NULL) && (info->empty)) {
4950 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
4951 ctxt->sax->endElement(ctxt->userData, name);
4952 htmlnamePop(ctxt);
4953 }
4954 ctxt->instate = XML_PARSER_CONTENT;
4955#ifdef DEBUG_PUSH
4956 xmlGenericError(xmlGenericErrorContext,
4957 "HPP: entering CONTENT\n");
4958#endif
4959 break;
4960 }
4961 case XML_PARSER_CONTENT: {
4962 long cons;
4963 /*
4964 * Handle preparsed entities and charRef
4965 */
4966 if (ctxt->token != 0) {
4967 xmlChar chr[2] = { 0 , 0 } ;
4968
4969 chr[0] = (xmlChar) ctxt->token;
4970 htmlCheckParagraph(ctxt);
4971 if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
4972 ctxt->sax->characters(ctxt->userData, chr, 1);
4973 ctxt->token = 0;
4974 ctxt->checkIndex = 0;
4975 }
4976 if ((avail == 1) && (terminate)) {
4977 cur = in->cur[0];
4978 if ((cur != '<') && (cur != '&')) {
4979 if (ctxt->sax != NULL) {
4980 if (IS_BLANK_CH(cur)) {
4981 if (ctxt->sax->ignorableWhitespace != NULL)
4982 ctxt->sax->ignorableWhitespace(
4983 ctxt->userData, &cur, 1);
4984 } else {
4985 htmlCheckParagraph(ctxt);
4986 if (ctxt->sax->characters != NULL)
4987 ctxt->sax->characters(
4988 ctxt->userData, &cur, 1);
4989 }
4990 }
4991 ctxt->token = 0;
4992 ctxt->checkIndex = 0;
4993 in->cur++;
4994 break;
4995 }
4996 }
4997 if (avail < 2)
4998 goto done;
4999 cur = in->cur[0];
5000 next = in->cur[1];
5001 cons = ctxt->nbChars;
5002 if ((xmlStrEqual(ctxt->name, BAD_CAST"script")) ||
5003 (xmlStrEqual(ctxt->name, BAD_CAST"style"))) {
5004 /*
5005 * Handle SCRIPT/STYLE separately
5006 */
5007 if (!terminate) {
5008 int idx;
5009 xmlChar val;
5010
5011 idx = htmlParseLookupSequence(ctxt, '<', '/', 0, 0);
5012 if (idx < 0)
5013 goto done;
5014 val = in->cur[idx + 2];
5015 if (val == 0) /* bad cut of input */
5016 goto done;
5017 }
5018 htmlParseScript(ctxt);
5019 if ((cur == '<') && (next == '/')) {
5020 ctxt->instate = XML_PARSER_END_TAG;
5021 ctxt->checkIndex = 0;
5022#ifdef DEBUG_PUSH
5023 xmlGenericError(xmlGenericErrorContext,
5024 "HPP: entering END_TAG\n");
5025#endif
5026 break;
5027 }
5028 } else {
5029 /*
5030 * Sometimes DOCTYPE arrives in the middle of the document
5031 */
5032 if ((cur == '<') && (next == '!') &&
5033 (UPP(2) == 'D') && (UPP(3) == 'O') &&
5034 (UPP(4) == 'C') && (UPP(5) == 'T') &&
5035 (UPP(6) == 'Y') && (UPP(7) == 'P') &&
5036 (UPP(8) == 'E')) {
5037 if ((!terminate) &&
5038 (htmlParseLookupSequence(ctxt, '>', 0, 0, 0) < 0))
5039 goto done;
5040 htmlParseErr(ctxt, XML_HTML_STRUCURE_ERROR,
5041 "Misplaced DOCTYPE declaration\n",
5042 BAD_CAST "DOCTYPE" , NULL);
5043 htmlParseDocTypeDecl(ctxt);
5044 } else if ((cur == '<') && (next == '!') &&
5045 (in->cur[2] == '-') && (in->cur[3] == '-')) {
5046 if ((!terminate) &&
5047 (htmlParseLookupSequence(
5048 ctxt, '-', '-', '>', 1) < 0))
5049 goto done;
5050#ifdef DEBUG_PUSH
5051 xmlGenericError(xmlGenericErrorContext,
5052 "HPP: Parsing Comment\n");
5053#endif
5054 htmlParseComment(ctxt);
5055 ctxt->instate = XML_PARSER_CONTENT;
5056 } else if ((cur == '<') && (next == '?')) {
5057 if ((!terminate) &&
5058 (htmlParseLookupSequence(ctxt, '>', 0, 0, 0) < 0))
5059 goto done;
5060#ifdef DEBUG_PUSH
5061 xmlGenericError(xmlGenericErrorContext,
5062 "HPP: Parsing PI\n");
5063#endif
5064 htmlParsePI(ctxt);
5065 ctxt->instate = XML_PARSER_CONTENT;
5066 } else if ((cur == '<') && (next == '!') && (avail < 4)) {
5067 goto done;
5068 } else if ((cur == '<') && (next == '/')) {
5069 ctxt->instate = XML_PARSER_END_TAG;
5070 ctxt->checkIndex = 0;
5071#ifdef DEBUG_PUSH
5072 xmlGenericError(xmlGenericErrorContext,
5073 "HPP: entering END_TAG\n");
5074#endif
5075 break;
5076 } else if (cur == '<') {
5077 ctxt->instate = XML_PARSER_START_TAG;
5078 ctxt->checkIndex = 0;
5079#ifdef DEBUG_PUSH
5080 xmlGenericError(xmlGenericErrorContext,
5081 "HPP: entering START_TAG\n");
5082#endif
5083 break;
5084 } else if (cur == '&') {
5085 if ((!terminate) &&
5086 (htmlParseLookupSequence(ctxt, ';', 0, 0, 0) < 0))
5087 goto done;
5088#ifdef DEBUG_PUSH
5089 xmlGenericError(xmlGenericErrorContext,
5090 "HPP: Parsing Reference\n");
5091#endif
5092 /* TODO: check generation of subtrees if noent !!! */
5093 htmlParseReference(ctxt);
5094 } else {
5095 /*
5096 * check that the text sequence is complete
5097 * before handing out the data to the parser
5098 * to avoid problems with erroneous end of
5099 * data detection.
5100 */
5101 if ((!terminate) &&
5102 (htmlParseLookupSequence(ctxt, '<', 0, 0, 0) < 0))
5103 goto done;
5104 ctxt->checkIndex = 0;
5105#ifdef DEBUG_PUSH
5106 xmlGenericError(xmlGenericErrorContext,
5107 "HPP: Parsing char data\n");
5108#endif
5109 htmlParseCharData(ctxt);
5110 }
5111 }
5112 if (cons == ctxt->nbChars) {
5113 if (ctxt->node != NULL) {
5114 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
5115 "detected an error in element content\n",
5116 NULL, NULL);
5117 }
5118 NEXT;
5119 break;
5120 }
5121
5122 break;
5123 }
5124 case XML_PARSER_END_TAG:
5125 if (avail < 2)
5126 goto done;
5127 if ((!terminate) &&
5128 (htmlParseLookupSequence(ctxt, '>', 0, 0, 0) < 0))
5129 goto done;
5130 htmlParseEndTag(ctxt);
5131 if (ctxt->nameNr == 0) {
5132 ctxt->instate = XML_PARSER_EPILOG;
5133 } else {
5134 ctxt->instate = XML_PARSER_CONTENT;
5135 }
5136 ctxt->checkIndex = 0;
5137#ifdef DEBUG_PUSH
5138 xmlGenericError(xmlGenericErrorContext,
5139 "HPP: entering CONTENT\n");
5140#endif
5141 break;
5142 case XML_PARSER_CDATA_SECTION:
5143 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
5144 "HPP: internal error, state == CDATA\n",
5145 NULL, NULL);
5146 ctxt->instate = XML_PARSER_CONTENT;
5147 ctxt->checkIndex = 0;
5148#ifdef DEBUG_PUSH
5149 xmlGenericError(xmlGenericErrorContext,
5150 "HPP: entering CONTENT\n");
5151#endif
5152 break;
5153 case XML_PARSER_DTD:
5154 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
5155 "HPP: internal error, state == DTD\n",
5156 NULL, NULL);
5157 ctxt->instate = XML_PARSER_CONTENT;
5158 ctxt->checkIndex = 0;
5159#ifdef DEBUG_PUSH
5160 xmlGenericError(xmlGenericErrorContext,
5161 "HPP: entering CONTENT\n");
5162#endif
5163 break;
5164 case XML_PARSER_COMMENT:
5165 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
5166 "HPP: internal error, state == COMMENT\n",
5167 NULL, NULL);
5168 ctxt->instate = XML_PARSER_CONTENT;
5169 ctxt->checkIndex = 0;
5170#ifdef DEBUG_PUSH
5171 xmlGenericError(xmlGenericErrorContext,
5172 "HPP: entering CONTENT\n");
5173#endif
5174 break;
5175 case XML_PARSER_PI:
5176 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
5177 "HPP: internal error, state == PI\n",
5178 NULL, NULL);
5179 ctxt->instate = XML_PARSER_CONTENT;
5180 ctxt->checkIndex = 0;
5181#ifdef DEBUG_PUSH
5182 xmlGenericError(xmlGenericErrorContext,
5183 "HPP: entering CONTENT\n");
5184#endif
5185 break;
5186 case XML_PARSER_ENTITY_DECL:
5187 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
5188 "HPP: internal error, state == ENTITY_DECL\n",
5189 NULL, NULL);
5190 ctxt->instate = XML_PARSER_CONTENT;
5191 ctxt->checkIndex = 0;
5192#ifdef DEBUG_PUSH
5193 xmlGenericError(xmlGenericErrorContext,
5194 "HPP: entering CONTENT\n");
5195#endif
5196 break;
5197 case XML_PARSER_ENTITY_VALUE:
5198 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
5199 "HPP: internal error, state == ENTITY_VALUE\n",
5200 NULL, NULL);
5201 ctxt->instate = XML_PARSER_CONTENT;
5202 ctxt->checkIndex = 0;
5203#ifdef DEBUG_PUSH
5204 xmlGenericError(xmlGenericErrorContext,
5205 "HPP: entering DTD\n");
5206#endif
5207 break;
5208 case XML_PARSER_ATTRIBUTE_VALUE:
5209 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
5210 "HPP: internal error, state == ATTRIBUTE_VALUE\n",
5211 NULL, NULL);
5212 ctxt->instate = XML_PARSER_START_TAG;
5213 ctxt->checkIndex = 0;
5214#ifdef DEBUG_PUSH
5215 xmlGenericError(xmlGenericErrorContext,
5216 "HPP: entering START_TAG\n");
5217#endif
5218 break;
5219 case XML_PARSER_SYSTEM_LITERAL:
5220 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
5221 "HPP: internal error, state == XML_PARSER_SYSTEM_LITERAL\n",
5222 NULL, NULL);
5223 ctxt->instate = XML_PARSER_CONTENT;
5224 ctxt->checkIndex = 0;
5225#ifdef DEBUG_PUSH
5226 xmlGenericError(xmlGenericErrorContext,
5227 "HPP: entering CONTENT\n");
5228#endif
5229 break;
5230 case XML_PARSER_IGNORE:
5231 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
5232 "HPP: internal error, state == XML_PARSER_IGNORE\n",
5233 NULL, NULL);
5234 ctxt->instate = XML_PARSER_CONTENT;
5235 ctxt->checkIndex = 0;
5236#ifdef DEBUG_PUSH
5237 xmlGenericError(xmlGenericErrorContext,
5238 "HPP: entering CONTENT\n");
5239#endif
5240 break;
5241 case XML_PARSER_PUBLIC_LITERAL:
5242 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
5243 "HPP: internal error, state == XML_PARSER_LITERAL\n",
5244 NULL, NULL);
5245 ctxt->instate = XML_PARSER_CONTENT;
5246 ctxt->checkIndex = 0;
5247#ifdef DEBUG_PUSH
5248 xmlGenericError(xmlGenericErrorContext,
5249 "HPP: entering CONTENT\n");
5250#endif
5251 break;
5252
5253 }
5254 }
5255done:
5256 if ((avail == 0) && (terminate)) {
5257 htmlAutoCloseOnEnd(ctxt);
5258 if ((ctxt->nameNr == 0) && (ctxt->instate != XML_PARSER_EOF)) {
5259 /*
5260 * SAX: end of the document processing.
5261 */
5262 ctxt->instate = XML_PARSER_EOF;
5263 if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
5264 ctxt->sax->endDocument(ctxt->userData);
5265 }
5266 }
5267 if ((ctxt->myDoc != NULL) &&
5268 ((terminate) || (ctxt->instate == XML_PARSER_EOF) ||
5269 (ctxt->instate == XML_PARSER_EPILOG))) {
5270 xmlDtdPtr dtd;
5271 dtd = xmlGetIntSubset(ctxt->myDoc);
5272 if (dtd == NULL)
5273 ctxt->myDoc->intSubset =
5274 xmlCreateIntSubset(ctxt->myDoc, BAD_CAST "html",
5275 BAD_CAST "-//W3C//DTD HTML 4.0 Transitional//EN",
5276 BAD_CAST "http://www.w3.org/TR/REC-html40/loose.dtd");
5277 }
5278#ifdef DEBUG_PUSH
5279 xmlGenericError(xmlGenericErrorContext, "HPP: done %d\n", ret);
5280#endif
5281 return(ret);
5282}
5283
5284/**
5285 * htmlParseChunk:
5286 * @ctxt: an HTML parser context
5287 * @chunk: an char array
5288 * @size: the size in byte of the chunk
5289 * @terminate: last chunk indicator
5290 *
5291 * Parse a Chunk of memory
5292 *
5293 * Returns zero if no error, the xmlParserErrors otherwise.
5294 */
5295int
5296htmlParseChunk(htmlParserCtxtPtr ctxt, const char *chunk, int size,
5297 int terminate) {
5298 if ((ctxt == NULL) || (ctxt->input == NULL)) {
5299 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
5300 "htmlParseChunk: context error\n", NULL, NULL);
5301 return(XML_ERR_INTERNAL_ERROR);
5302 }
5303 if ((size > 0) && (chunk != NULL) && (ctxt->input != NULL) &&
5304 (ctxt->input->buf != NULL) && (ctxt->instate != XML_PARSER_EOF)) {
5305 int base = ctxt->input->base - ctxt->input->buf->buffer->content;
5306 int cur = ctxt->input->cur - ctxt->input->base;
5307 int res;
5308
5309 res = xmlParserInputBufferPush(ctxt->input->buf, size, chunk);
5310 if (res < 0) {
5311 ctxt->errNo = XML_PARSER_EOF;
5312 ctxt->disableSAX = 1;
5313 return (XML_PARSER_EOF);
5314 }
5315 ctxt->input->base = ctxt->input->buf->buffer->content + base;
5316 ctxt->input->cur = ctxt->input->base + cur;
5317 ctxt->input->end =
5318 &ctxt->input->buf->buffer->content[ctxt->input->buf->buffer->use];
5319#ifdef DEBUG_PUSH
5320 xmlGenericError(xmlGenericErrorContext, "HPP: pushed %d\n", size);
5321#endif
5322
5323#if 0
5324 if ((terminate) || (ctxt->input->buf->buffer->use > 80))
5325 htmlParseTryOrFinish(ctxt, terminate);
5326#endif
5327 } else if (ctxt->instate != XML_PARSER_EOF) {
5328 if ((ctxt->input != NULL) && ctxt->input->buf != NULL) {
5329 xmlParserInputBufferPtr in = ctxt->input->buf;
5330 if ((in->encoder != NULL) && (in->buffer != NULL) &&
5331 (in->raw != NULL)) {
5332 int nbchars;
5333
5334 nbchars = xmlCharEncInFunc(in->encoder, in->buffer, in->raw);
5335 if (nbchars < 0) {
5336 htmlParseErr(ctxt, XML_ERR_INVALID_ENCODING,
5337 "encoder error\n", NULL, NULL);
5338 return(XML_ERR_INVALID_ENCODING);
5339 }
5340 }
5341 }
5342 }
5343 htmlParseTryOrFinish(ctxt, terminate);
5344 if (terminate) {
5345 if ((ctxt->instate != XML_PARSER_EOF) &&
5346 (ctxt->instate != XML_PARSER_EPILOG) &&
5347 (ctxt->instate != XML_PARSER_MISC)) {
5348 ctxt->errNo = XML_ERR_DOCUMENT_END;
5349 ctxt->wellFormed = 0;
5350 }
5351 if (ctxt->instate != XML_PARSER_EOF) {
5352 if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
5353 ctxt->sax->endDocument(ctxt->userData);
5354 }
5355 ctxt->instate = XML_PARSER_EOF;
5356 }
5357 return((xmlParserErrors) ctxt->errNo);
5358}
5359
5360/************************************************************************
5361 * *
5362 * User entry points *
5363 * *
5364 ************************************************************************/
5365
5366/**
5367 * htmlCreatePushParserCtxt:
5368 * @sax: a SAX handler
5369 * @user_data: The user data returned on SAX callbacks
5370 * @chunk: a pointer to an array of chars
5371 * @size: number of chars in the array
5372 * @filename: an optional file name or URI
5373 * @enc: an optional encoding
5374 *
5375 * Create a parser context for using the HTML parser in push mode
5376 * The value of @filename is used for fetching external entities
5377 * and error/warning reports.
5378 *
5379 * Returns the new parser context or NULL
5380 */
5381htmlParserCtxtPtr
5382htmlCreatePushParserCtxt(htmlSAXHandlerPtr sax, void *user_data,
5383 const char *chunk, int size, const char *filename,
5384 xmlCharEncoding enc) {
5385 htmlParserCtxtPtr ctxt;
5386 htmlParserInputPtr inputStream;
5387 xmlParserInputBufferPtr buf;
5388
5389 xmlInitParser();
5390
5391 buf = xmlAllocParserInputBuffer(enc);
5392 if (buf == NULL) return(NULL);
5393
5394 ctxt = htmlNewParserCtxt();
5395 if (ctxt == NULL) {
5396 xmlFreeParserInputBuffer(buf);
5397 return(NULL);
5398 }
5399 if(enc==XML_CHAR_ENCODING_UTF8 || buf->encoder)
5400 ctxt->charset=XML_CHAR_ENCODING_UTF8;
5401 if (sax != NULL) {
5402 if (ctxt->sax != (xmlSAXHandlerPtr) &htmlDefaultSAXHandler)
5403 xmlFree(ctxt->sax);
5404 ctxt->sax = (htmlSAXHandlerPtr) xmlMalloc(sizeof(htmlSAXHandler));
5405 if (ctxt->sax == NULL) {
5406 xmlFree(buf);
5407 xmlFree(ctxt);
5408 return(NULL);
5409 }
5410 memcpy(ctxt->sax, sax, sizeof(htmlSAXHandler));
5411 if (user_data != NULL)
5412 ctxt->userData = user_data;
5413 }
5414 if (filename == NULL) {
5415 ctxt->directory = NULL;
5416 } else {
5417 ctxt->directory = xmlParserGetDirectory(filename);
5418 }
5419
5420 inputStream = htmlNewInputStream(ctxt);
5421 if (inputStream == NULL) {
5422 xmlFreeParserCtxt(ctxt);
5423 xmlFree(buf);
5424 return(NULL);
5425 }
5426
5427 if (filename == NULL)
5428 inputStream->filename = NULL;
5429 else
5430 inputStream->filename = (char *)
5431 xmlCanonicPath((const xmlChar *) filename);
5432 inputStream->buf = buf;
5433 inputStream->base = inputStream->buf->buffer->content;
5434 inputStream->cur = inputStream->buf->buffer->content;
5435 inputStream->end =
5436 &inputStream->buf->buffer->content[inputStream->buf->buffer->use];
5437
5438 inputPush(ctxt, inputStream);
5439
5440 if ((size > 0) && (chunk != NULL) && (ctxt->input != NULL) &&
5441 (ctxt->input->buf != NULL)) {
5442 int base = ctxt->input->base - ctxt->input->buf->buffer->content;
5443 int cur = ctxt->input->cur - ctxt->input->base;
5444
5445 xmlParserInputBufferPush(ctxt->input->buf, size, chunk);
5446
5447 ctxt->input->base = ctxt->input->buf->buffer->content + base;
5448 ctxt->input->cur = ctxt->input->base + cur;
5449 ctxt->input->end =
5450 &ctxt->input->buf->buffer->content[ctxt->input->buf->buffer->use];
5451#ifdef DEBUG_PUSH
5452 xmlGenericError(xmlGenericErrorContext, "HPP: pushed %d\n", size);
5453#endif
5454 }
5455 ctxt->progressive = 1;
5456
5457 return(ctxt);
5458}
5459#endif /* LIBXML_PUSH_ENABLED */
5460
5461/**
5462 * htmlSAXParseDoc:
5463 * @cur: a pointer to an array of xmlChar
5464 * @encoding: a free form C string describing the HTML document encoding, or NULL
5465 * @sax: the SAX handler block
5466 * @userData: if using SAX, this pointer will be provided on callbacks.
5467 *
5468 * Parse an HTML in-memory document. If sax is not NULL, use the SAX callbacks
5469 * to handle parse events. If sax is NULL, fallback to the default DOM
5470 * behavior and return a tree.
5471 *
5472 * Returns the resulting document tree unless SAX is NULL or the document is
5473 * not well formed.
5474 */
5475
5476htmlDocPtr
5477htmlSAXParseDoc(xmlChar *cur, const char *encoding, htmlSAXHandlerPtr sax, void *userData) {
5478 htmlDocPtr ret;
5479 htmlParserCtxtPtr ctxt;
5480
5481 xmlInitParser();
5482
5483 if (cur == NULL) return(NULL);
5484
5485
5486 ctxt = htmlCreateDocParserCtxt(cur, encoding);
5487 if (ctxt == NULL) return(NULL);
5488 if (sax != NULL) {
5489 if (ctxt->sax != NULL) xmlFree (ctxt->sax);
5490 ctxt->sax = sax;
5491 ctxt->userData = userData;
5492 }
5493
5494 htmlParseDocument(ctxt);
5495 ret = ctxt->myDoc;
5496 if (sax != NULL) {
5497 ctxt->sax = NULL;
5498 ctxt->userData = NULL;
5499 }
5500 htmlFreeParserCtxt(ctxt);
5501
5502 return(ret);
5503}
5504
5505/**
5506 * htmlParseDoc:
5507 * @cur: a pointer to an array of xmlChar
5508 * @encoding: a free form C string describing the HTML document encoding, or NULL
5509 *
5510 * parse an HTML in-memory document and build a tree.
5511 *
5512 * Returns the resulting document tree
5513 */
5514
5515htmlDocPtr
5516htmlParseDoc(xmlChar *cur, const char *encoding) {
5517 return(htmlSAXParseDoc(cur, encoding, NULL, NULL));
5518}
5519
5520
5521/**
5522 * htmlCreateFileParserCtxt:
5523 * @filename: the filename
5524 * @encoding: a free form C string describing the HTML document encoding, or NULL
5525 *
5526 * Create a parser context for a file content.
5527 * Automatic support for ZLIB/Compress compressed document is provided
5528 * by default if found at compile-time.
5529 *
5530 * Returns the new parser context or NULL
5531 */
5532htmlParserCtxtPtr
5533htmlCreateFileParserCtxt(const char *filename, const char *encoding)
5534{
5535 htmlParserCtxtPtr ctxt;
5536 htmlParserInputPtr inputStream;
5537 char *canonicFilename;
5538 /* htmlCharEncoding enc; */
5539 xmlChar *content, *content_line = (xmlChar *) "charset=";
5540
5541 if (filename == NULL)
5542 return(NULL);
5543
5544 ctxt = htmlNewParserCtxt();
5545 if (ctxt == NULL) {
5546 return(NULL);
5547 }
5548 canonicFilename = (char *) xmlCanonicPath((const xmlChar *) filename);
5549 if (canonicFilename == NULL) {
5550#ifdef LIBXML_SAX1_ENABLED
5551 if (xmlDefaultSAXHandler.error != NULL) {
5552 xmlDefaultSAXHandler.error(NULL, "out of memory\n");
5553 }
5554#endif
5555 xmlFreeParserCtxt(ctxt);
5556 return(NULL);
5557 }
5558
5559 inputStream = xmlLoadExternalEntity(canonicFilename, NULL, ctxt);
5560 xmlFree(canonicFilename);
5561 if (inputStream == NULL) {
5562 xmlFreeParserCtxt(ctxt);
5563 return(NULL);
5564 }
5565
5566 inputPush(ctxt, inputStream);
5567
5568 /* set encoding */
5569 if (encoding) {
5570 content = xmlMallocAtomic (xmlStrlen(content_line) + strlen(encoding) + 1);
5571 if (content) {
5572 strcpy ((char *)content, (char *)content_line);
5573 strcat ((char *)content, (char *)encoding);
5574 htmlCheckEncoding (ctxt, content);
5575 xmlFree (content);
5576 }
5577 }
5578
5579 return(ctxt);
5580}
5581
5582/**
5583 * htmlSAXParseFile:
5584 * @filename: the filename
5585 * @encoding: a free form C string describing the HTML document encoding, or NULL
5586 * @sax: the SAX handler block
5587 * @userData: if using SAX, this pointer will be provided on callbacks.
5588 *
5589 * parse an HTML file and build a tree. Automatic support for ZLIB/Compress
5590 * compressed document is provided by default if found at compile-time.
5591 * It use the given SAX function block to handle the parsing callback.
5592 * If sax is NULL, fallback to the default DOM tree building routines.
5593 *
5594 * Returns the resulting document tree unless SAX is NULL or the document is
5595 * not well formed.
5596 */
5597
5598htmlDocPtr
5599htmlSAXParseFile(const char *filename, const char *encoding, htmlSAXHandlerPtr sax,
5600 void *userData) {
5601 htmlDocPtr ret;
5602 htmlParserCtxtPtr ctxt;
5603 htmlSAXHandlerPtr oldsax = NULL;
5604
5605 xmlInitParser();
5606
5607 ctxt = htmlCreateFileParserCtxt(filename, encoding);
5608 if (ctxt == NULL) return(NULL);
5609 if (sax != NULL) {
5610 oldsax = ctxt->sax;
5611 ctxt->sax = sax;
5612 ctxt->userData = userData;
5613 }
5614
5615 htmlParseDocument(ctxt);
5616
5617 ret = ctxt->myDoc;
5618 if (sax != NULL) {
5619 ctxt->sax = oldsax;
5620 ctxt->userData = NULL;
5621 }
5622 htmlFreeParserCtxt(ctxt);
5623
5624 return(ret);
5625}
5626
5627/**
5628 * htmlParseFile:
5629 * @filename: the filename
5630 * @encoding: a free form C string describing the HTML document encoding, or NULL
5631 *
5632 * parse an HTML file and build a tree. Automatic support for ZLIB/Compress
5633 * compressed document is provided by default if found at compile-time.
5634 *
5635 * Returns the resulting document tree
5636 */
5637
5638htmlDocPtr
5639htmlParseFile(const char *filename, const char *encoding) {
5640 return(htmlSAXParseFile(filename, encoding, NULL, NULL));
5641}
5642
5643/**
5644 * htmlHandleOmittedElem:
5645 * @val: int 0 or 1
5646 *
5647 * Set and return the previous value for handling HTML omitted tags.
5648 *
5649 * Returns the last value for 0 for no handling, 1 for auto insertion.
5650 */
5651
5652int
5653htmlHandleOmittedElem(int val) {
5654 int old = htmlOmittedDefaultValue;
5655
5656 htmlOmittedDefaultValue = val;
5657 return(old);
5658}
5659
5660/**
5661 * htmlElementAllowedHere:
5662 * @parent: HTML parent element
5663 * @elt: HTML element
5664 *
5665 * Checks whether an HTML element may be a direct child of a parent element.
5666 * Note - doesn't check for deprecated elements
5667 *
5668 * Returns 1 if allowed; 0 otherwise.
5669 */
5670int
5671htmlElementAllowedHere(const htmlElemDesc* parent, const xmlChar* elt) {
5672 const char** p ;
5673
5674 if ( ! elt || ! parent || ! parent->subelts )
5675 return 0 ;
5676
5677 for ( p = parent->subelts; *p; ++p )
5678 if ( !xmlStrcmp((const xmlChar *)*p, elt) )
5679 return 1 ;
5680
5681 return 0 ;
5682}
5683/**
5684 * htmlElementStatusHere:
5685 * @parent: HTML parent element
5686 * @elt: HTML element
5687 *
5688 * Checks whether an HTML element may be a direct child of a parent element.
5689 * and if so whether it is valid or deprecated.
5690 *
5691 * Returns one of HTML_VALID, HTML_DEPRECATED, HTML_INVALID
5692 */
5693htmlStatus
5694htmlElementStatusHere(const htmlElemDesc* parent, const htmlElemDesc* elt) {
5695 if ( ! parent || ! elt )
5696 return HTML_INVALID ;
5697 if ( ! htmlElementAllowedHere(parent, (const xmlChar*) elt->name ) )
5698 return HTML_INVALID ;
5699
5700 return ( elt->dtd == 0 ) ? HTML_VALID : HTML_DEPRECATED ;
5701}
5702/**
5703 * htmlAttrAllowed:
5704 * @elt: HTML element
5705 * @attr: HTML attribute
5706 * @legacy: whether to allow deprecated attributes
5707 *
5708 * Checks whether an attribute is valid for an element
5709 * Has full knowledge of Required and Deprecated attributes
5710 *
5711 * Returns one of HTML_REQUIRED, HTML_VALID, HTML_DEPRECATED, HTML_INVALID
5712 */
5713htmlStatus
5714htmlAttrAllowed(const htmlElemDesc* elt, const xmlChar* attr, int legacy) {
5715 const char** p ;
5716
5717 if ( !elt || ! attr )
5718 return HTML_INVALID ;
5719
5720 if ( elt->attrs_req )
5721 for ( p = elt->attrs_req; *p; ++p)
5722 if ( !xmlStrcmp((const xmlChar*)*p, attr) )
5723 return HTML_REQUIRED ;
5724
5725 if ( elt->attrs_opt )
5726 for ( p = elt->attrs_opt; *p; ++p)
5727 if ( !xmlStrcmp((const xmlChar*)*p, attr) )
5728 return HTML_VALID ;
5729
5730 if ( legacy && elt->attrs_depr )
5731 for ( p = elt->attrs_depr; *p; ++p)
5732 if ( !xmlStrcmp((const xmlChar*)*p, attr) )
5733 return HTML_DEPRECATED ;
5734
5735 return HTML_INVALID ;
5736}
5737/**
5738 * htmlNodeStatus:
5739 * @node: an htmlNodePtr in a tree
5740 * @legacy: whether to allow deprecated elements (YES is faster here
5741 * for Element nodes)
5742 *
5743 * Checks whether the tree node is valid. Experimental (the author
5744 * only uses the HTML enhancements in a SAX parser)
5745 *
5746 * Return: for Element nodes, a return from htmlElementAllowedHere (if
5747 * legacy allowed) or htmlElementStatusHere (otherwise).
5748 * for Attribute nodes, a return from htmlAttrAllowed
5749 * for other nodes, HTML_NA (no checks performed)
5750 */
5751htmlStatus
5752htmlNodeStatus(const htmlNodePtr node, int legacy) {
5753 if ( ! node )
5754 return HTML_INVALID ;
5755
5756 switch ( node->type ) {
5757 case XML_ELEMENT_NODE:
5758 return legacy
5759 ? ( htmlElementAllowedHere (
5760 htmlTagLookup(node->parent->name) , node->name
5761 ) ? HTML_VALID : HTML_INVALID )
5762 : htmlElementStatusHere(
5763 htmlTagLookup(node->parent->name) ,
5764 htmlTagLookup(node->name) )
5765 ;
5766 case XML_ATTRIBUTE_NODE:
5767 return htmlAttrAllowed(
5768 htmlTagLookup(node->parent->name) , node->name, legacy) ;
5769 default: return HTML_NA ;
5770 }
5771}
5772/************************************************************************
5773 * *
5774 * New set (2.6.0) of simpler and more flexible APIs *
5775 * *
5776 ************************************************************************/
5777/**
5778 * DICT_FREE:
5779 * @str: a string
5780 *
5781 * Free a string if it is not owned by the "dict" dictionnary in the
5782 * current scope
5783 */
5784#define DICT_FREE(str) \
5785 if ((str) && ((!dict) || \
5786 (xmlDictOwns(dict, (const xmlChar *)(str)) == 0))) \
5787 xmlFree((char *)(str));
5788
5789/**
5790 * htmlCtxtReset:
5791 * @ctxt: an HTML parser context
5792 *
5793 * Reset a parser context
5794 */
5795void
5796htmlCtxtReset(htmlParserCtxtPtr ctxt)
5797{
5798 xmlParserInputPtr input;
5799 xmlDictPtr dict;
5800
5801 if (ctxt == NULL)
5802 return;
5803
5804 xmlInitParser();
5805 dict = ctxt->dict;
5806
5807 while ((input = inputPop(ctxt)) != NULL) { /* Non consuming */
5808 xmlFreeInputStream(input);
5809 }
5810 ctxt->inputNr = 0;
5811 ctxt->input = NULL;
5812
5813 ctxt->spaceNr = 0;
5814 if (ctxt->spaceTab != NULL) {
5815 ctxt->spaceTab[0] = -1;
5816 ctxt->space = &ctxt->spaceTab[0];
5817 } else {
5818 ctxt->space = NULL;
5819 }
5820
5821
5822 ctxt->nodeNr = 0;
5823 ctxt->node = NULL;
5824
5825 ctxt->nameNr = 0;
5826 ctxt->name = NULL;
5827
5828 DICT_FREE(ctxt->version);
5829 ctxt->version = NULL;
5830 DICT_FREE(ctxt->encoding);
5831 ctxt->encoding = NULL;
5832 DICT_FREE(ctxt->directory);
5833 ctxt->directory = NULL;
5834 DICT_FREE(ctxt->extSubURI);
5835 ctxt->extSubURI = NULL;
5836 DICT_FREE(ctxt->extSubSystem);
5837 ctxt->extSubSystem = NULL;
5838 if (ctxt->myDoc != NULL)
5839 xmlFreeDoc(ctxt->myDoc);
5840 ctxt->myDoc = NULL;
5841
5842 ctxt->standalone = -1;
5843 ctxt->hasExternalSubset = 0;
5844 ctxt->hasPErefs = 0;
5845 ctxt->html = 1;
5846 ctxt->external = 0;
5847 ctxt->instate = XML_PARSER_START;
5848 ctxt->token = 0;
5849
5850 ctxt->wellFormed = 1;
5851 ctxt->nsWellFormed = 1;
5852 ctxt->valid = 1;
5853 ctxt->vctxt.userData = ctxt;
5854 ctxt->vctxt.error = xmlParserValidityError;
5855 ctxt->vctxt.warning = xmlParserValidityWarning;
5856 ctxt->record_info = 0;
5857 ctxt->nbChars = 0;
5858 ctxt->checkIndex = 0;
5859 ctxt->inSubset = 0;
5860 ctxt->errNo = XML_ERR_OK;
5861 ctxt->depth = 0;
5862 ctxt->charset = XML_CHAR_ENCODING_NONE;
5863 ctxt->catalogs = NULL;
5864 xmlInitNodeInfoSeq(&ctxt->node_seq);
5865
5866 if (ctxt->attsDefault != NULL) {
5867 xmlHashFree(ctxt->attsDefault, (xmlHashDeallocator) xmlFree);
5868 ctxt->attsDefault = NULL;
5869 }
5870 if (ctxt->attsSpecial != NULL) {
5871 xmlHashFree(ctxt->attsSpecial, NULL);
5872 ctxt->attsSpecial = NULL;
5873 }
5874}
5875
5876/**
5877 * htmlCtxtUseOptions:
5878 * @ctxt: an HTML parser context
5879 * @options: a combination of htmlParserOption(s)
5880 *
5881 * Applies the options to the parser context
5882 *
5883 * Returns 0 in case of success, the set of unknown or unimplemented options
5884 * in case of error.
5885 */
5886int
5887htmlCtxtUseOptions(htmlParserCtxtPtr ctxt, int options)
5888{
5889 if (ctxt == NULL)
5890 return(-1);
5891
5892 if (options & HTML_PARSE_NOWARNING) {
5893 ctxt->sax->warning = NULL;
5894 ctxt->vctxt.warning = NULL;
5895 options -= XML_PARSE_NOWARNING;
5896 ctxt->options |= XML_PARSE_NOWARNING;
5897 }
5898 if (options & HTML_PARSE_NOERROR) {
5899 ctxt->sax->error = NULL;
5900 ctxt->vctxt.error = NULL;
5901 ctxt->sax->fatalError = NULL;
5902 options -= XML_PARSE_NOERROR;
5903 ctxt->options |= XML_PARSE_NOERROR;
5904 }
5905 if (options & HTML_PARSE_PEDANTIC) {
5906 ctxt->pedantic = 1;
5907 options -= XML_PARSE_PEDANTIC;
5908 ctxt->options |= XML_PARSE_PEDANTIC;
5909 } else
5910 ctxt->pedantic = 0;
5911 if (options & XML_PARSE_NOBLANKS) {
5912 ctxt->keepBlanks = 0;
5913 ctxt->sax->ignorableWhitespace = xmlSAX2IgnorableWhitespace;
5914 options -= XML_PARSE_NOBLANKS;
5915 ctxt->options |= XML_PARSE_NOBLANKS;
5916 } else
5917 ctxt->keepBlanks = 1;
5918 if (options & HTML_PARSE_RECOVER) {
5919 ctxt->recovery = 1;
5920 options -= HTML_PARSE_RECOVER;
5921 } else
5922 ctxt->recovery = 0;
5923 if (options & HTML_PARSE_COMPACT) {
5924 ctxt->options |= HTML_PARSE_COMPACT;
5925 options -= HTML_PARSE_COMPACT;
5926 }
5927 ctxt->dictNames = 0;
5928 return (options);
5929}
5930
5931/**
5932 * htmlDoRead:
5933 * @ctxt: an HTML parser context
5934 * @URL: the base URL to use for the document
5935 * @encoding: the document encoding, or NULL
5936 * @options: a combination of htmlParserOption(s)
5937 * @reuse: keep the context for reuse
5938 *
5939 * Common front-end for the htmlRead functions
5940 *
5941 * Returns the resulting document tree or NULL
5942 */
5943static htmlDocPtr
5944htmlDoRead(htmlParserCtxtPtr ctxt, const char *URL, const char *encoding,
5945 int options, int reuse)
5946{
5947 htmlDocPtr ret;
5948
5949 htmlCtxtUseOptions(ctxt, options);
5950 ctxt->html = 1;
5951 if (encoding != NULL) {
5952 xmlCharEncodingHandlerPtr hdlr;
5953
5954 hdlr = xmlFindCharEncodingHandler(encoding);
5955 if (hdlr != NULL)
5956 xmlSwitchToEncoding(ctxt, hdlr);
5957 }
5958 if ((URL != NULL) && (ctxt->input != NULL) &&
5959 (ctxt->input->filename == NULL))
5960 ctxt->input->filename = (char *) xmlStrdup((const xmlChar *) URL);
5961 htmlParseDocument(ctxt);
5962 ret = ctxt->myDoc;
5963 ctxt->myDoc = NULL;
5964 if (!reuse) {
5965 if ((ctxt->dictNames) &&
5966 (ret != NULL) &&
5967 (ret->dict == ctxt->dict))
5968 ctxt->dict = NULL;
5969 xmlFreeParserCtxt(ctxt);
5970 }
5971 return (ret);
5972}
5973
5974/**
5975 * htmlReadDoc:
5976 * @cur: a pointer to a zero terminated string
5977 * @URL: the base URL to use for the document
5978 * @encoding: the document encoding, or NULL
5979 * @options: a combination of htmlParserOption(s)
5980 *
5981 * parse an XML in-memory document and build a tree.
5982 *
5983 * Returns the resulting document tree
5984 */
5985htmlDocPtr
5986htmlReadDoc(const xmlChar * cur, const char *URL, const char *encoding, int options)
5987{
5988 htmlParserCtxtPtr ctxt;
5989
5990 if (cur == NULL)
5991 return (NULL);
5992
5993 xmlInitParser();
5994 ctxt = htmlCreateDocParserCtxt(cur, NULL);
5995 if (ctxt == NULL)
5996 return (NULL);
5997 return (htmlDoRead(ctxt, URL, encoding, options, 0));
5998}
5999
6000/**
6001 * htmlReadFile:
6002 * @filename: a file or URL
6003 * @encoding: the document encoding, or NULL
6004 * @options: a combination of htmlParserOption(s)
6005 *
6006 * parse an XML file from the filesystem or the network.
6007 *
6008 * Returns the resulting document tree
6009 */
6010htmlDocPtr
6011htmlReadFile(const char *filename, const char *encoding, int options)
6012{
6013 htmlParserCtxtPtr ctxt;
6014
6015 xmlInitParser();
6016 ctxt = htmlCreateFileParserCtxt(filename, encoding);
6017 if (ctxt == NULL)
6018 return (NULL);
6019 return (htmlDoRead(ctxt, NULL, NULL, options, 0));
6020}
6021
6022/**
6023 * htmlReadMemory:
6024 * @buffer: a pointer to a char array
6025 * @size: the size of the array
6026 * @URL: the base URL to use for the document
6027 * @encoding: the document encoding, or NULL
6028 * @options: a combination of htmlParserOption(s)
6029 *
6030 * parse an XML in-memory document and build a tree.
6031 *
6032 * Returns the resulting document tree
6033 */
6034htmlDocPtr
6035htmlReadMemory(const char *buffer, int size, const char *URL, const char *encoding, int options)
6036{
6037 htmlParserCtxtPtr ctxt;
6038
6039 xmlInitParser();
6040 ctxt = xmlCreateMemoryParserCtxt(buffer, size);
6041 if (ctxt == NULL)
6042 return (NULL);
6043 htmlDefaultSAXHandlerInit();
6044 if (ctxt->sax != NULL)
6045 memcpy(ctxt->sax, &htmlDefaultSAXHandler, sizeof(xmlSAXHandlerV1));
6046 return (htmlDoRead(ctxt, URL, encoding, options, 0));
6047}
6048
6049/**
6050 * htmlReadFd:
6051 * @fd: an open file descriptor
6052 * @URL: the base URL to use for the document
6053 * @encoding: the document encoding, or NULL
6054 * @options: a combination of htmlParserOption(s)
6055 *
6056 * parse an XML from a file descriptor and build a tree.
6057 *
6058 * Returns the resulting document tree
6059 */
6060htmlDocPtr
6061htmlReadFd(int fd, const char *URL, const char *encoding, int options)
6062{
6063 htmlParserCtxtPtr ctxt;
6064 xmlParserInputBufferPtr input;
6065 xmlParserInputPtr stream;
6066
6067 if (fd < 0)
6068 return (NULL);
6069
6070 xmlInitParser();
6071 input = xmlParserInputBufferCreateFd(fd, XML_CHAR_ENCODING_NONE);
6072 if (input == NULL)
6073 return (NULL);
6074 ctxt = xmlNewParserCtxt();
6075 if (ctxt == NULL) {
6076 xmlFreeParserInputBuffer(input);
6077 return (NULL);
6078 }
6079 stream = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE);
6080 if (stream == NULL) {
6081 xmlFreeParserInputBuffer(input);
6082 xmlFreeParserCtxt(ctxt);
6083 return (NULL);
6084 }
6085 inputPush(ctxt, stream);
6086 return (htmlDoRead(ctxt, URL, encoding, options, 0));
6087}
6088
6089/**
6090 * htmlReadIO:
6091 * @ioread: an I/O read function
6092 * @ioclose: an I/O close function
6093 * @ioctx: an I/O handler
6094 * @URL: the base URL to use for the document
6095 * @encoding: the document encoding, or NULL
6096 * @options: a combination of htmlParserOption(s)
6097 *
6098 * parse an HTML document from I/O functions and source and build a tree.
6099 *
6100 * Returns the resulting document tree
6101 */
6102htmlDocPtr
6103htmlReadIO(xmlInputReadCallback ioread, xmlInputCloseCallback ioclose,
6104 void *ioctx, const char *URL, const char *encoding, int options)
6105{
6106 htmlParserCtxtPtr ctxt;
6107 xmlParserInputBufferPtr input;
6108 xmlParserInputPtr stream;
6109
6110 if (ioread == NULL)
6111 return (NULL);
6112 xmlInitParser();
6113
6114 input = xmlParserInputBufferCreateIO(ioread, ioclose, ioctx,
6115 XML_CHAR_ENCODING_NONE);
6116 if (input == NULL)
6117 return (NULL);
6118 ctxt = htmlNewParserCtxt();
6119 if (ctxt == NULL) {
6120 xmlFreeParserInputBuffer(input);
6121 return (NULL);
6122 }
6123 stream = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE);
6124 if (stream == NULL) {
6125 xmlFreeParserInputBuffer(input);
6126 xmlFreeParserCtxt(ctxt);
6127 return (NULL);
6128 }
6129 inputPush(ctxt, stream);
6130 return (htmlDoRead(ctxt, URL, encoding, options, 0));
6131}
6132
6133/**
6134 * htmlCtxtReadDoc:
6135 * @ctxt: an HTML parser context
6136 * @cur: a pointer to a zero terminated string
6137 * @URL: the base URL to use for the document
6138 * @encoding: the document encoding, or NULL
6139 * @options: a combination of htmlParserOption(s)
6140 *
6141 * parse an XML in-memory document and build a tree.
6142 * This reuses the existing @ctxt parser context
6143 *
6144 * Returns the resulting document tree
6145 */
6146htmlDocPtr
6147htmlCtxtReadDoc(htmlParserCtxtPtr ctxt, const xmlChar * cur,
6148 const char *URL, const char *encoding, int options)
6149{
6150 xmlParserInputPtr stream;
6151
6152 if (cur == NULL)
6153 return (NULL);
6154 if (ctxt == NULL)
6155 return (NULL);
6156
6157 htmlCtxtReset(ctxt);
6158
6159 stream = xmlNewStringInputStream(ctxt, cur);
6160 if (stream == NULL) {
6161 return (NULL);
6162 }
6163 inputPush(ctxt, stream);
6164 return (htmlDoRead(ctxt, URL, encoding, options, 1));
6165}
6166
6167/**
6168 * htmlCtxtReadFile:
6169 * @ctxt: an HTML parser context
6170 * @filename: a file or URL
6171 * @encoding: the document encoding, or NULL
6172 * @options: a combination of htmlParserOption(s)
6173 *
6174 * parse an XML file from the filesystem or the network.
6175 * This reuses the existing @ctxt parser context
6176 *
6177 * Returns the resulting document tree
6178 */
6179htmlDocPtr
6180htmlCtxtReadFile(htmlParserCtxtPtr ctxt, const char *filename,
6181 const char *encoding, int options)
6182{
6183 xmlParserInputPtr stream;
6184
6185 if (filename == NULL)
6186 return (NULL);
6187 if (ctxt == NULL)
6188 return (NULL);
6189
6190 htmlCtxtReset(ctxt);
6191
6192 stream = xmlLoadExternalEntity(filename, NULL, ctxt);
6193 if (stream == NULL) {
6194 return (NULL);
6195 }
6196 inputPush(ctxt, stream);
6197 return (htmlDoRead(ctxt, NULL, encoding, options, 1));
6198}
6199
6200/**
6201 * htmlCtxtReadMemory:
6202 * @ctxt: an HTML parser context
6203 * @buffer: a pointer to a char array
6204 * @size: the size of the array
6205 * @URL: the base URL to use for the document
6206 * @encoding: the document encoding, or NULL
6207 * @options: a combination of htmlParserOption(s)
6208 *
6209 * parse an XML in-memory document and build a tree.
6210 * This reuses the existing @ctxt parser context
6211 *
6212 * Returns the resulting document tree
6213 */
6214htmlDocPtr
6215htmlCtxtReadMemory(htmlParserCtxtPtr ctxt, const char *buffer, int size,
6216 const char *URL, const char *encoding, int options)
6217{
6218 xmlParserInputBufferPtr input;
6219 xmlParserInputPtr stream;
6220
6221 if (ctxt == NULL)
6222 return (NULL);
6223 if (buffer == NULL)
6224 return (NULL);
6225
6226 htmlCtxtReset(ctxt);
6227
6228 input = xmlParserInputBufferCreateMem(buffer, size, XML_CHAR_ENCODING_NONE);
6229 if (input == NULL) {
6230 return(NULL);
6231 }
6232
6233 stream = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE);
6234 if (stream == NULL) {
6235 xmlFreeParserInputBuffer(input);
6236 return(NULL);
6237 }
6238
6239 inputPush(ctxt, stream);
6240 return (htmlDoRead(ctxt, URL, encoding, options, 1));
6241}
6242
6243/**
6244 * htmlCtxtReadFd:
6245 * @ctxt: an HTML parser context
6246 * @fd: an open file descriptor
6247 * @URL: the base URL to use for the document
6248 * @encoding: the document encoding, or NULL
6249 * @options: a combination of htmlParserOption(s)
6250 *
6251 * parse an XML from a file descriptor and build a tree.
6252 * This reuses the existing @ctxt parser context
6253 *
6254 * Returns the resulting document tree
6255 */
6256htmlDocPtr
6257htmlCtxtReadFd(htmlParserCtxtPtr ctxt, int fd,
6258 const char *URL, const char *encoding, int options)
6259{
6260 xmlParserInputBufferPtr input;
6261 xmlParserInputPtr stream;
6262
6263 if (fd < 0)
6264 return (NULL);
6265 if (ctxt == NULL)
6266 return (NULL);
6267
6268 htmlCtxtReset(ctxt);
6269
6270
6271 input = xmlParserInputBufferCreateFd(fd, XML_CHAR_ENCODING_NONE);
6272 if (input == NULL)
6273 return (NULL);
6274 stream = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE);
6275 if (stream == NULL) {
6276 xmlFreeParserInputBuffer(input);
6277 return (NULL);
6278 }
6279 inputPush(ctxt, stream);
6280 return (htmlDoRead(ctxt, URL, encoding, options, 1));
6281}
6282
6283/**
6284 * htmlCtxtReadIO:
6285 * @ctxt: an HTML parser context
6286 * @ioread: an I/O read function
6287 * @ioclose: an I/O close function
6288 * @ioctx: an I/O handler
6289 * @URL: the base URL to use for the document
6290 * @encoding: the document encoding, or NULL
6291 * @options: a combination of htmlParserOption(s)
6292 *
6293 * parse an HTML document from I/O functions and source and build a tree.
6294 * This reuses the existing @ctxt parser context
6295 *
6296 * Returns the resulting document tree
6297 */
6298htmlDocPtr
6299htmlCtxtReadIO(htmlParserCtxtPtr ctxt, xmlInputReadCallback ioread,
6300 xmlInputCloseCallback ioclose, void *ioctx,
6301 const char *URL,
6302 const char *encoding, int options)
6303{
6304 xmlParserInputBufferPtr input;
6305 xmlParserInputPtr stream;
6306
6307 if (ioread == NULL)
6308 return (NULL);
6309 if (ctxt == NULL)
6310 return (NULL);
6311
6312 htmlCtxtReset(ctxt);
6313
6314 input = xmlParserInputBufferCreateIO(ioread, ioclose, ioctx,
6315 XML_CHAR_ENCODING_NONE);
6316 if (input == NULL)
6317 return (NULL);
6318 stream = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE);
6319 if (stream == NULL) {
6320 xmlFreeParserInputBuffer(input);
6321 return (NULL);
6322 }
6323 inputPush(ctxt, stream);
6324 return (htmlDoRead(ctxt, URL, encoding, options, 1));
6325}
6326
6327#define bottom_HTMLparser
6328#include "elfgcchack.h"
6329#endif /* LIBXML_HTML_ENABLED */
Note: See TracBrowser for help on using the repository browser.

© 2023 Oracle
Contact – Privacy policy – Terms of Use