2 * HTMLparser.c : an HTML 4.0 non-verifying parser
4 * See Copyright for the status of this software.
11 #ifdef LIBXML_HTML_ENABLED
20 #ifdef HAVE_SYS_STAT_H
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>
47 #define HTML_MAX_NAMELEN 1000
48 #define HTML_PARSER_BIG_BUFFER_SIZE 1000
49 #define HTML_PARSER_BUFFER_SIZE 100
52 /* #define DEBUG_PUSH */
54 static int htmlOmittedDefaultValue = 1;
56 xmlChar * htmlDecodeEntities(htmlParserCtxtPtr ctxt, int len,
57 xmlChar end, xmlChar end2, xmlChar end3);
58 static void htmlParseComment(htmlParserCtxtPtr ctxt);
60 /************************************************************************
62 * Some factorized error routines *
64 ************************************************************************/
68 * @ctxt: an HTML parser context
69 * @extra: extra informations
71 * Handle a redefinition of attribute error
74 htmlErrMemory(xmlParserCtxtPtr ctxt, const char *extra)
76 if ((ctxt != NULL) && (ctxt->disableSAX != 0) &&
77 (ctxt->instate == XML_PARSER_EOF))
80 ctxt->errNo = XML_ERR_NO_MEMORY;
81 ctxt->instate = XML_PARSER_EOF;
85 __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER,
86 XML_ERR_NO_MEMORY, XML_ERR_FATAL, NULL, 0, extra,
88 "Memory allocation failed : %s\n", extra);
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");
97 * @ctxt: an HTML parser context
98 * @error: the error number
99 * @msg: the error message
100 * @str1: string infor
101 * @str2: string infor
103 * Handle a fatal parser error, i.e. violating Well-Formedness constraints
106 htmlParseErr(xmlParserCtxtPtr ctxt, xmlParserErrors error,
107 const char *msg, const xmlChar *str1, const xmlChar *str2)
109 if ((ctxt != NULL) && (ctxt->disableSAX != 0) &&
110 (ctxt->instate == XML_PARSER_EOF))
114 __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_HTML, error,
115 XML_ERR_ERROR, NULL, 0,
116 (const char *) str1, (const char *) str2,
120 ctxt->wellFormed = 0;
125 * @ctxt: an HTML parser context
126 * @error: the error number
127 * @msg: the error message
130 * Handle a fatal parser error, i.e. violating Well-Formedness constraints
133 htmlParseErrInt(xmlParserCtxtPtr ctxt, xmlParserErrors error,
134 const char *msg, int val)
136 if ((ctxt != NULL) && (ctxt->disableSAX != 0) &&
137 (ctxt->instate == XML_PARSER_EOF))
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);
145 ctxt->wellFormed = 0;
148 /************************************************************************
150 * Parser stacks related functions and macros *
152 ************************************************************************/
156 * @ctxt: an HTML parser context
157 * @value: the element name
159 * Pushes a new element name on top of the name stack
161 * Returns 0 in case of error, the index in the stack otherwise
164 htmlnamePush(htmlParserCtxtPtr ctxt, const xmlChar * value)
166 if ((ctxt->html < 3) && (xmlStrEqual(value, BAD_CAST "head")))
168 if ((ctxt->html < 10) && (xmlStrEqual(value, BAD_CAST "body")))
170 if (ctxt->nameNr >= ctxt->nameMax) {
172 ctxt->nameTab = (const xmlChar * *)
173 xmlRealloc((xmlChar * *)ctxt->nameTab,
175 sizeof(ctxt->nameTab[0]));
176 if (ctxt->nameTab == NULL) {
177 htmlErrMemory(ctxt, NULL);
181 ctxt->nameTab[ctxt->nameNr] = value;
183 return (ctxt->nameNr++);
187 * @ctxt: an HTML parser context
189 * Pops the top element name from the name stack
191 * Returns the name just removed
193 static const xmlChar *
194 htmlnamePop(htmlParserCtxtPtr ctxt)
198 if (ctxt->nameNr <= 0)
201 if (ctxt->nameNr < 0)
203 if (ctxt->nameNr > 0)
204 ctxt->name = ctxt->nameTab[ctxt->nameNr - 1];
207 ret = ctxt->nameTab[ctxt->nameNr];
208 ctxt->nameTab[ctxt->nameNr] = NULL;
214 * @ctxt: an HTML parser context
215 * @value: the node info
217 * Pushes a new element name on top of the node info stack
219 * Returns 0 in case of error, the index in the stack otherwise
222 htmlNodeInfoPush(htmlParserCtxtPtr ctxt, htmlParserNodeInfo *value)
224 if (ctxt->nodeInfoNr >= ctxt->nodeInfoMax) {
225 if (ctxt->nodeInfoMax == 0)
226 ctxt->nodeInfoMax = 5;
227 ctxt->nodeInfoMax *= 2;
228 ctxt->nodeInfoTab = (htmlParserNodeInfo *)
229 xmlRealloc((htmlParserNodeInfo *)ctxt->nodeInfoTab,
231 sizeof(ctxt->nodeInfoTab[0]));
232 if (ctxt->nodeInfoTab == NULL) {
233 htmlErrMemory(ctxt, NULL);
237 ctxt->nodeInfoTab[ctxt->nodeInfoNr] = *value;
238 ctxt->nodeInfo = &ctxt->nodeInfoTab[ctxt->nodeInfoNr];
239 return (ctxt->nodeInfoNr++);
244 * @ctxt: an HTML parser context
246 * Pops the top element name from the node info stack
248 * Returns 0 in case of error, the pointer to NodeInfo otherwise
250 static htmlParserNodeInfo *
251 htmlNodeInfoPop(htmlParserCtxtPtr ctxt)
253 if (ctxt->nodeInfoNr <= 0)
256 if (ctxt->nodeInfoNr < 0)
258 if (ctxt->nodeInfoNr > 0)
259 ctxt->nodeInfo = &ctxt->nodeInfoTab[ctxt->nodeInfoNr - 1];
261 ctxt->nodeInfo = NULL;
262 return &ctxt->nodeInfoTab[ctxt->nodeInfoNr];
266 * Macros for accessing the content. Those should be used only by the parser,
269 * Dirty macros, i.e. one need to make assumption on the context to use them
271 * CUR_PTR return the current pointer to the xmlChar to be parsed.
272 * CUR returns the current xmlChar value, i.e. a 8 bit value if compiled
273 * in ISO-Latin or UTF-8, and the current 16 bit value if compiled
274 * in UNICODE mode. This should be used internally by the parser
275 * only to compare to ASCII values otherwise it would break when
276 * running with UTF-8 encoding.
277 * NXT(n) returns the n'th next xmlChar. Same as CUR is should be used only
278 * to compare on ASCII based substring.
279 * UPP(n) returns the n'th next xmlChar converted to uppercase. Same as CUR
280 * it should be used only to compare on ASCII based substring.
281 * SKIP(n) Skip n xmlChar, and must also be used only to skip ASCII defined
282 * strings without newlines within the parser.
284 * Clean macros, not dependent of an ASCII context, expect UTF-8 encoding
286 * CURRENT Returns the current char value, with the full decoding of
287 * UTF-8 if we are using this mode. It returns an int.
288 * NEXT Skip to the next character, this does the proper decoding
289 * in UTF-8 mode. It also pop-up unfinished entities on the fly.
290 * NEXTL(l) Skip the current unicode character of l xmlChars long.
291 * COPY(to) copy one char to *to, increment CUR_PTR and to accordingly
294 #define UPPER (toupper(*ctxt->input->cur))
296 #define SKIP(val) ctxt->nbChars += (val),ctxt->input->cur += (val),ctxt->input->col+=(val)
298 #define NXT(val) ctxt->input->cur[(val)]
300 #define UPP(val) (toupper(ctxt->input->cur[(val)]))
302 #define CUR_PTR ctxt->input->cur
304 #define SHRINK if ((ctxt->input->cur - ctxt->input->base > 2 * INPUT_CHUNK) && \
305 (ctxt->input->end - ctxt->input->cur < 2 * INPUT_CHUNK)) \
306 xmlParserInputShrink(ctxt->input)
308 #define GROW if ((ctxt->progressive == 0) && \
309 (ctxt->input->end - ctxt->input->cur < INPUT_CHUNK)) \
310 xmlParserInputGrow(ctxt->input, INPUT_CHUNK)
312 #define CURRENT ((int) (*ctxt->input->cur))
314 #define SKIP_BLANKS htmlSkipBlankChars(ctxt)
316 /* Inported from XML */
318 /* #define CUR (ctxt->token ? ctxt->token : (int) (*ctxt->input->cur)) */
319 #define CUR ((int) (*ctxt->input->cur))
320 #define NEXT xmlNextChar(ctxt)
322 #define RAW (ctxt->token ? -1 : (*ctxt->input->cur))
325 #define NEXTL(l) do { \
326 if (*(ctxt->input->cur) == '\n') { \
327 ctxt->input->line++; ctxt->input->col = 1; \
328 } else ctxt->input->col++; \
329 ctxt->token = 0; ctxt->input->cur += l; ctxt->nbChars++; \
334 if (*ctxt->input->cur == '%') xmlParserHandlePEReference(ctxt); \
335 if (*ctxt->input->cur == '&') xmlParserHandleReference(ctxt);
338 #define CUR_CHAR(l) htmlCurrentChar(ctxt, &l)
339 #define CUR_SCHAR(s, l) xmlStringCurrentChar(ctxt, s, &l)
341 #define COPY_BUF(l,b,i,v) \
342 if (l == 1) b[i++] = (xmlChar) v; \
343 else i += xmlCopyChar(l,&b[i],v)
347 * @the HTML parser context
349 * Ty to find and encoding in the current data available in the input
350 * buffer this is needed to try to switch to the proper encoding when
351 * one face a character error.
352 * That's an heuristic, since it's operating outside of parsing it could
353 * try to use a meta which had been commented out, that's the reason it
354 * should only be used in case of error, not as a default.
356 * Returns an encoding string or NULL if not found, the string need to
360 htmlFindEncoding(xmlParserCtxtPtr ctxt) {
361 const xmlChar *start, *cur, *end;
363 if ((ctxt == NULL) || (ctxt->input == NULL) ||
364 (ctxt->input->encoding != NULL) || (ctxt->input->buf == NULL) ||
365 (ctxt->input->buf->encoder != NULL))
367 if ((ctxt->input->cur == NULL) || (ctxt->input->end == NULL))
370 start = ctxt->input->cur;
371 end = ctxt->input->end;
372 /* we also expect the input buffer to be zero terminated */
376 cur = xmlStrcasestr(start, BAD_CAST "HTTP-EQUIV");
379 cur = xmlStrcasestr(cur, BAD_CAST "CONTENT");
382 cur = xmlStrcasestr(cur, BAD_CAST "CHARSET=");
387 while (((*cur >= 'A') && (*cur <= 'Z')) ||
388 ((*cur >= 'a') && (*cur <= 'z')) ||
389 ((*cur >= '0') && (*cur <= '9')) ||
390 (*cur == '-') || (*cur == '_') || (*cur == ':') || (*cur == '/'))
394 return(xmlStrndup(start, cur - start));
399 * @ctxt: the HTML parser context
400 * @len: pointer to the length of the char read
402 * The current char value, if using UTF-8 this may actually span multiple
403 * bytes in the input buffer. Implement the end of line normalization:
404 * 2.11 End-of-Line Handling
405 * If the encoding is unspecified, in the case we find an ISO-Latin-1
406 * char, then the encoding converter is plugged in automatically.
408 * Returns the current char value and its length
412 htmlCurrentChar(xmlParserCtxtPtr ctxt, int *len) {
413 if (ctxt->instate == XML_PARSER_EOF)
416 if (ctxt->token != 0) {
420 if (ctxt->charset == XML_CHAR_ENCODING_UTF8) {
422 * We are supposed to handle UTF8, check it's valid
423 * From rfc2044: encoding of the Unicode values on UTF-8:
425 * UCS-4 range (hex.) UTF-8 octet sequence (binary)
426 * 0000 0000-0000 007F 0xxxxxxx
427 * 0000 0080-0000 07FF 110xxxxx 10xxxxxx
428 * 0000 0800-0000 FFFF 1110xxxx 10xxxxxx 10xxxxxx
430 * Check for the 0x110000 limit too
432 const unsigned char *cur = ctxt->input->cur;
439 xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
440 cur = ctxt->input->cur;
442 if ((cur[1] & 0xc0) != 0x80)
444 if ((c & 0xe0) == 0xe0) {
447 xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
448 cur = ctxt->input->cur;
450 if ((cur[2] & 0xc0) != 0x80)
452 if ((c & 0xf0) == 0xf0) {
454 xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
455 cur = ctxt->input->cur;
457 if (((c & 0xf8) != 0xf0) ||
458 ((cur[3] & 0xc0) != 0x80))
462 val = (cur[0] & 0x7) << 18;
463 val |= (cur[1] & 0x3f) << 12;
464 val |= (cur[2] & 0x3f) << 6;
465 val |= cur[3] & 0x3f;
469 val = (cur[0] & 0xf) << 12;
470 val |= (cur[1] & 0x3f) << 6;
471 val |= cur[2] & 0x3f;
476 val = (cur[0] & 0x1f) << 6;
477 val |= cur[1] & 0x3f;
480 htmlParseErrInt(ctxt, XML_ERR_INVALID_CHAR,
481 "Char 0x%X out of allowed range\n", val);
485 if ((*ctxt->input->cur == 0) &&
486 (ctxt->input->cur < ctxt->input->end)) {
487 htmlParseErrInt(ctxt, XML_ERR_INVALID_CHAR,
488 "Char 0x%X out of allowed range\n", 0);
494 return((int) *ctxt->input->cur);
498 * Assume it's a fixed length encoding (1) with
499 * a compatible encoding for the ASCII set, since
500 * XML constructs only use < 128 chars
503 if ((int) *ctxt->input->cur < 0x80)
504 return((int) *ctxt->input->cur);
507 * Humm this is bad, do an automatic flow conversion
511 xmlCharEncodingHandlerPtr handler;
513 guess = htmlFindEncoding(ctxt);
515 xmlSwitchEncoding(ctxt, XML_CHAR_ENCODING_8859_1);
517 if (ctxt->input->encoding != NULL)
518 xmlFree((xmlChar *) ctxt->input->encoding);
519 ctxt->input->encoding = guess;
520 handler = xmlFindCharEncodingHandler((const char *) guess);
521 if (handler != NULL) {
522 xmlSwitchToEncoding(ctxt, handler);
524 htmlParseErr(ctxt, XML_ERR_INVALID_ENCODING,
525 "Unsupported encoding %s", guess, NULL);
528 ctxt->charset = XML_CHAR_ENCODING_UTF8;
531 return(xmlCurrentChar(ctxt, len));
535 * If we detect an UTF8 error that probably mean that the
536 * input encoding didn't get properly advertized in the
537 * declaration header. Report the error and switch the encoding
538 * to ISO-Latin-1 (if you don't like this policy, just declare the
544 if (ctxt->input->end - ctxt->input->cur >= 4) {
545 snprintf(buffer, 149, "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n",
546 ctxt->input->cur[0], ctxt->input->cur[1],
547 ctxt->input->cur[2], ctxt->input->cur[3]);
549 snprintf(buffer, 149, "Bytes: 0x%02X\n", ctxt->input->cur[0]);
551 htmlParseErr(ctxt, XML_ERR_INVALID_ENCODING,
552 "Input is not proper UTF-8, indicate encoding !\n",
553 BAD_CAST buffer, NULL);
556 ctxt->charset = XML_CHAR_ENCODING_8859_1;
558 return((int) *ctxt->input->cur);
562 * htmlSkipBlankChars:
563 * @ctxt: the HTML parser context
565 * skip all blanks character found at that point in the input streams.
567 * Returns the number of space chars skipped
571 htmlSkipBlankChars(xmlParserCtxtPtr ctxt) {
574 while (IS_BLANK_CH(*(ctxt->input->cur))) {
575 if ((*ctxt->input->cur == 0) &&
576 (xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0)) {
579 if (*(ctxt->input->cur) == '\n') {
580 ctxt->input->line++; ctxt->input->col = 1;
581 } else ctxt->input->col++;
584 if (*ctxt->input->cur == 0)
585 xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
594 /************************************************************************
596 * The list of HTML elements and their properties *
598 ************************************************************************/
601 * Start Tag: 1 means the start tag can be ommited
602 * End Tag: 1 means the end tag can be ommited
603 * 2 means it's forbidden (empty elements)
604 * 3 means the tag is stylistic and should be closed easily
605 * Depr: this element is deprecated
606 * DTD: 1 means that this element is valid only in the Loose DTD
607 * 2 means that this element is valid only in the Frameset DTD
609 * Name,Start Tag,End Tag,Save End,Empty,Deprecated,DTD,inline,Description
610 , subElements , impliedsubelt , Attributes, userdata
613 /* Definitions and a couple of vars for HTML Elements */
615 #define FONTSTYLE "tt", "i", "b", "u", "s", "strike", "big", "small"
616 #define NB_FONTSTYLE 8
617 #define PHRASE "em", "strong", "dfn", "code", "samp", "kbd", "var", "cite", "abbr", "acronym"
619 #define SPECIAL "a", "img", "applet", "embed", "object", "font", "basefont", "br", "script", "map", "q", "sub", "sup", "span", "bdo", "iframe"
620 #define NB_SPECIAL 16
621 #define INLINE FONTSTYLE, PHRASE, SPECIAL, FORMCTRL
622 #define NB_INLINE NB_PCDATA + NB_FONTSTYLE + NB_PHRASE + NB_SPECIAL + NB_FORMCTRL
623 #define BLOCK HEADING, LIST, "pre", "p", "dl", "div", "center", "noscript", "noframes", "blockquote", "form", "isindex", "hr", "table", "fieldset", "address"
624 #define NB_BLOCK NB_HEADING + NB_LIST + 14
625 #define FORMCTRL "input", "select", "textarea", "label", "button"
626 #define NB_FORMCTRL 5
629 #define HEADING "h1", "h2", "h3", "h4", "h5", "h6"
631 #define LIST "ul", "ol", "dir", "menu"
634 #define NB_MODIFIER 0
635 #define FLOW BLOCK,INLINE
636 #define NB_FLOW NB_BLOCK + NB_INLINE
640 static const char* const html_flow[] = { FLOW, NULL } ;
641 static const char* const html_inline[] = { INLINE, NULL } ;
643 /* placeholders: elts with content but no subelements */
644 static const char* const html_pcdata[] = { NULL } ;
645 #define html_cdata html_pcdata
648 /* ... and for HTML Attributes */
650 #define COREATTRS "id", "class", "style", "title"
651 #define NB_COREATTRS 4
652 #define I18N "lang", "dir"
654 #define EVENTS "onclick", "ondblclick", "onmousedown", "onmouseup", "onmouseover", "onmouseout", "onkeypress", "onkeydown", "onkeyup"
656 #define ATTRS COREATTRS,I18N,EVENTS
657 #define NB_ATTRS NB_NB_COREATTRS + NB_I18N + NB_EVENTS
658 #define CELLHALIGN "align", "char", "charoff"
659 #define NB_CELLHALIGN 3
660 #define CELLVALIGN "valign"
661 #define NB_CELLVALIGN 1
663 static const char* const html_attrs[] = { ATTRS, NULL } ;
664 static const char* const core_i18n_attrs[] = { COREATTRS, I18N, NULL } ;
665 static const char* const core_attrs[] = { COREATTRS, NULL } ;
666 static const char* const i18n_attrs[] = { I18N, NULL } ;
669 /* Other declarations that should go inline ... */
670 static const char* const a_attrs[] = { ATTRS, "charset", "type", "name",
671 "href", "hreflang", "rel", "rev", "accesskey", "shape", "coords",
672 "tabindex", "onfocus", "onblur", NULL } ;
673 static const char* const target_attr[] = { "target", NULL } ;
674 static const char* const rows_cols_attr[] = { "rows", "cols", NULL } ;
675 static const char* const alt_attr[] = { "alt", NULL } ;
676 static const char* const src_alt_attrs[] = { "src", "alt", NULL } ;
677 static const char* const href_attrs[] = { "href", NULL } ;
678 static const char* const clear_attrs[] = { "clear", NULL } ;
679 static const char* const inline_p[] = { INLINE, "p", NULL } ;
681 static const char* const flow_param[] = { FLOW, "param", NULL } ;
682 static const char* const applet_attrs[] = { COREATTRS , "codebase",
683 "archive", "alt", "name", "height", "width", "align",
684 "hspace", "vspace", NULL } ;
685 static const char* const area_attrs[] = { "shape", "coords", "href", "nohref",
686 "tabindex", "accesskey", "onfocus", "onblur", NULL } ;
687 static const char* const basefont_attrs[] =
688 { "id", "size", "color", "face", NULL } ;
689 static const char* const quote_attrs[] = { ATTRS, "cite", NULL } ;
690 static const char* const body_contents[] = { FLOW, "ins", "del", NULL } ;
691 static const char* const body_attrs[] = { ATTRS, "onload", "onunload", NULL } ;
692 static const char* const body_depr[] = { "background", "bgcolor", "text",
693 "link", "vlink", "alink", NULL } ;
694 static const char* const button_attrs[] = { ATTRS, "name", "value", "type",
695 "disabled", "tabindex", "accesskey", "onfocus", "onblur", NULL } ;
698 static const char* const col_attrs[] = { ATTRS, "span", "width", CELLHALIGN, CELLVALIGN, NULL } ;
699 static const char* const col_elt[] = { "col", NULL } ;
700 static const char* const edit_attrs[] = { ATTRS, "datetime", "cite", NULL } ;
701 static const char* const compact_attrs[] = { ATTRS, "compact", NULL } ;
702 static const char* const dl_contents[] = { "dt", "dd", NULL } ;
703 static const char* const compact_attr[] = { "compact", NULL } ;
704 static const char* const label_attr[] = { "label", NULL } ;
705 static const char* const fieldset_contents[] = { FLOW, "legend" } ;
706 static const char* const font_attrs[] = { COREATTRS, I18N, "size", "color", "face" , NULL } ;
707 static const char* const form_contents[] = { HEADING, LIST, INLINE, "pre", "p", "div", "center", "noscript", "noframes", "blockquote", "isindex", "hr", "table", "fieldset", "address", NULL } ;
708 static const char* const form_attrs[] = { ATTRS, "method", "enctype", "accept", "name", "onsubmit", "onreset", "accept-charset", NULL } ;
709 static const char* const frame_attrs[] = { COREATTRS, "longdesc", "name", "src", "frameborder", "marginwidth", "marginheight", "noresize", "scrolling" , NULL } ;
710 static const char* const frameset_attrs[] = { COREATTRS, "rows", "cols", "onload", "onunload", NULL } ;
711 static const char* const frameset_contents[] = { "frameset", "frame", "noframes", NULL } ;
712 static const char* const head_attrs[] = { I18N, "profile", NULL } ;
713 static const char* const head_contents[] = { "title", "isindex", "base", "script", "style", "meta", "link", "object", NULL } ;
714 static const char* const hr_depr[] = { "align", "noshade", "size", "width", NULL } ;
715 static const char* const version_attr[] = { "version", NULL } ;
716 static const char* const html_content[] = { "head", "body", "frameset", NULL } ;
717 static const char* const iframe_attrs[] = { COREATTRS, "longdesc", "name", "src", "frameborder", "marginwidth", "marginheight", "scrolling", "align", "height", "width", NULL } ;
718 static const char* const img_attrs[] = { ATTRS, "longdesc", "name", "height", "width", "usemap", "ismap", NULL } ;
719 static 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 } ;
720 static 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 } ;
721 static const char* const prompt_attrs[] = { COREATTRS, I18N, "prompt", NULL } ;
722 static const char* const label_attrs[] = { ATTRS, "for", "accesskey", "onfocus", "onblur", NULL } ;
723 static const char* const legend_attrs[] = { ATTRS, "accesskey", NULL } ;
724 static const char* const align_attr[] = { "align", NULL } ;
725 static const char* const link_attrs[] = { ATTRS, "charset", "href", "hreflang", "type", "rel", "rev", "media", NULL } ;
726 static const char* const map_contents[] = { BLOCK, "area", NULL } ;
727 static const char* const name_attr[] = { "name", NULL } ;
728 static const char* const action_attr[] = { "action", NULL } ;
729 static const char* const blockli_elt[] = { BLOCK, "li", NULL } ;
730 static const char* const meta_attrs[] = { I18N, "http-equiv", "name", "scheme", NULL } ;
731 static const char* const content_attr[] = { "content", NULL } ;
732 static const char* const type_attr[] = { "type", NULL } ;
733 static const char* const noframes_content[] = { "body", FLOW MODIFIER, NULL } ;
734 static const char* const object_contents[] = { FLOW, "param", NULL } ;
735 static const char* const object_attrs[] = { ATTRS, "declare", "classid", "codebase", "data", "type", "codetype", "archive", "standby", "height", "width", "usemap", "name", "tabindex", NULL } ;
736 static const char* const object_depr[] = { "align", "border", "hspace", "vspace", NULL } ;
737 static const char* const ol_attrs[] = { "type", "compact", "start", NULL} ;
738 static const char* const option_elt[] = { "option", NULL } ;
739 static const char* const optgroup_attrs[] = { ATTRS, "disabled", NULL } ;
740 static const char* const option_attrs[] = { ATTRS, "disabled", "label", "selected", "value", NULL } ;
741 static const char* const param_attrs[] = { "id", "value", "valuetype", "type", NULL } ;
742 static const char* const width_attr[] = { "width", NULL } ;
743 static const char* const pre_content[] = { PHRASE, "tt", "i", "b", "u", "s", "strike", "a", "br", "script", "map", "q", "span", "bdo", "iframe", NULL } ;
744 static const char* const script_attrs[] = { "charset", "src", "defer", "event", "for", NULL } ;
745 static const char* const language_attr[] = { "language", NULL } ;
746 static const char* const select_content[] = { "optgroup", "option", NULL } ;
747 static const char* const select_attrs[] = { ATTRS, "name", "size", "multiple", "disabled", "tabindex", "onfocus", "onblur", "onchange", NULL } ;
748 static const char* const style_attrs[] = { I18N, "media", "title", NULL } ;
749 static const char* const table_attrs[] = { ATTRS, "summary", "width", "border", "frame", "rules", "cellspacing", "cellpadding", "datapagesize", NULL } ;
750 static const char* const table_depr[] = { "align", "bgcolor", NULL } ;
751 static const char* const table_contents[] = { "caption", "col", "colgroup", "thead", "tfoot", "tbody", "tr", NULL} ;
752 static const char* const tr_elt[] = { "tr", NULL } ;
753 static const char* const talign_attrs[] = { ATTRS, CELLHALIGN, CELLVALIGN, NULL} ;
754 static const char* const th_td_depr[] = { "nowrap", "bgcolor", "width", "height", NULL } ;
755 static const char* const th_td_attr[] = { ATTRS, "abbr", "axis", "headers", "scope", "rowspan", "colspan", CELLHALIGN, CELLVALIGN, NULL } ;
756 static const char* const textarea_attrs[] = { ATTRS, "name", "disabled", "readonly", "tabindex", "accesskey", "onfocus", "onblur", "onselect", "onchange", NULL } ;
757 static const char* const tr_contents[] = { "th", "td", NULL } ;
758 static const char* const bgcolor_attr[] = { "bgcolor", NULL } ;
759 static const char* const li_elt[] = { "li", NULL } ;
760 static const char* const ul_depr[] = { "type", "compact", NULL} ;
761 static const char* const dir_attr[] = { "dir", NULL} ;
763 #define DECL (const char**)
765 static const htmlElemDesc
766 html40ElementTable[] = {
767 { "a", 0, 0, 0, 0, 0, 0, 1, "anchor ",
768 DECL html_inline , NULL , DECL a_attrs , DECL target_attr, NULL
770 { "abbr", 0, 0, 0, 0, 0, 0, 1, "abbreviated form",
771 DECL html_inline , NULL , DECL html_attrs, NULL, NULL
773 { "acronym", 0, 0, 0, 0, 0, 0, 1, "",
774 DECL html_inline , NULL , DECL html_attrs, NULL, NULL
776 { "address", 0, 0, 0, 0, 0, 0, 0, "information on author ",
777 DECL inline_p , NULL , DECL html_attrs, NULL, NULL
779 { "applet", 0, 0, 0, 0, 1, 1, 2, "java applet ",
780 DECL flow_param , NULL , NULL , DECL applet_attrs, NULL
782 { "area", 0, 2, 2, 1, 0, 0, 0, "client-side image map area ",
783 EMPTY , NULL , DECL area_attrs , DECL target_attr, DECL alt_attr
785 { "b", 0, 3, 0, 0, 0, 0, 1, "bold text style",
786 DECL html_inline , NULL , DECL html_attrs, NULL, NULL
788 { "base", 0, 2, 2, 1, 0, 0, 0, "document base uri ",
789 EMPTY , NULL , NULL , DECL target_attr, DECL href_attrs
791 { "basefont", 0, 2, 2, 1, 1, 1, 1, "base font size " ,
792 EMPTY , NULL , NULL, DECL basefont_attrs, NULL
794 { "bdo", 0, 0, 0, 0, 0, 0, 1, "i18n bidi over-ride ",
795 DECL html_inline , NULL , DECL core_i18n_attrs, NULL, DECL dir_attr
797 { "big", 0, 3, 0, 0, 0, 0, 1, "large text style",
798 DECL html_inline , NULL , DECL html_attrs, NULL, NULL
800 { "blockquote", 0, 0, 0, 0, 0, 0, 0, "long quotation ",
801 DECL html_flow , NULL , DECL quote_attrs , NULL, NULL
803 { "body", 1, 1, 0, 0, 0, 0, 0, "document body ",
804 DECL body_contents , "div" , DECL body_attrs, DECL body_depr, NULL
806 { "br", 0, 2, 2, 1, 0, 0, 1, "forced line break ",
807 EMPTY , NULL , DECL core_attrs, DECL clear_attrs , NULL
809 { "button", 0, 0, 0, 0, 0, 0, 2, "push button ",
810 DECL html_flow MODIFIER , NULL , DECL button_attrs, NULL, NULL
812 { "caption", 0, 0, 0, 0, 0, 0, 0, "table caption ",
813 DECL html_inline , NULL , DECL html_attrs, NULL, NULL
815 { "center", 0, 3, 0, 0, 1, 1, 0, "shorthand for div align=center ",
816 DECL html_flow , NULL , NULL, DECL html_attrs, NULL
818 { "cite", 0, 0, 0, 0, 0, 0, 1, "citation",
819 DECL html_inline , NULL , DECL html_attrs, NULL, NULL
821 { "code", 0, 0, 0, 0, 0, 0, 1, "computer code fragment",
822 DECL html_inline , NULL , DECL html_attrs, NULL, NULL
824 { "col", 0, 2, 2, 1, 0, 0, 0, "table column ",
825 EMPTY , NULL , DECL col_attrs , NULL, NULL
827 { "colgroup", 0, 1, 0, 0, 0, 0, 0, "table column group ",
828 DECL col_elt , "col" , DECL col_attrs , NULL, NULL
830 { "dd", 0, 1, 0, 0, 0, 0, 0, "definition description ",
831 DECL html_flow , NULL , DECL html_attrs, NULL, NULL
833 { "del", 0, 0, 0, 0, 0, 0, 2, "deleted text ",
834 DECL html_flow , NULL , DECL edit_attrs , NULL, NULL
836 { "dfn", 0, 0, 0, 0, 0, 0, 1, "instance definition",
837 DECL html_inline , NULL , DECL html_attrs, NULL, NULL
839 { "dir", 0, 0, 0, 0, 1, 1, 0, "directory list",
840 DECL blockli_elt, "li" , NULL, DECL compact_attrs, NULL
842 { "div", 0, 0, 0, 0, 0, 0, 0, "generic language/style container",
843 DECL html_flow, NULL, DECL html_attrs, DECL align_attr, NULL
845 { "dl", 0, 0, 0, 0, 0, 0, 0, "definition list ",
846 DECL dl_contents , "dd" , DECL html_attrs, DECL compact_attr, NULL
848 { "dt", 0, 1, 0, 0, 0, 0, 0, "definition term ",
849 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
851 { "em", 0, 3, 0, 0, 0, 0, 1, "emphasis",
852 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
854 { "embed", 0, 1, 0, 0, 1, 1, 1, "generic embedded object ",
855 EMPTY, NULL, DECL embed_attrs, NULL, NULL
857 { "fieldset", 0, 0, 0, 0, 0, 0, 0, "form control group ",
858 DECL fieldset_contents , NULL, DECL html_attrs, NULL, NULL
860 { "font", 0, 3, 0, 0, 1, 1, 1, "local change to font ",
861 DECL html_inline, NULL, NULL, DECL font_attrs, NULL
863 { "form", 0, 0, 0, 0, 0, 0, 0, "interactive form ",
864 DECL form_contents, "fieldset", DECL form_attrs , DECL target_attr, DECL action_attr
866 { "frame", 0, 2, 2, 1, 0, 2, 0, "subwindow " ,
867 EMPTY, NULL, NULL, DECL frame_attrs, NULL
869 { "frameset", 0, 0, 0, 0, 0, 2, 0, "window subdivision" ,
870 DECL frameset_contents, "noframes" , NULL , DECL frameset_attrs, NULL
872 { "h1", 0, 0, 0, 0, 0, 0, 0, "heading ",
873 DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
875 { "h2", 0, 0, 0, 0, 0, 0, 0, "heading ",
876 DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
878 { "h3", 0, 0, 0, 0, 0, 0, 0, "heading ",
879 DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
881 { "h4", 0, 0, 0, 0, 0, 0, 0, "heading ",
882 DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
884 { "h5", 0, 0, 0, 0, 0, 0, 0, "heading ",
885 DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
887 { "h6", 0, 0, 0, 0, 0, 0, 0, "heading ",
888 DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
890 { "head", 1, 1, 0, 0, 0, 0, 0, "document head ",
891 DECL head_contents, NULL, DECL head_attrs, NULL, NULL
893 { "hr", 0, 2, 2, 1, 0, 0, 0, "horizontal rule " ,
894 EMPTY, NULL, DECL html_attrs, DECL hr_depr, NULL
896 { "html", 1, 1, 0, 0, 0, 0, 0, "document root element ",
897 DECL html_content , NULL , DECL i18n_attrs, DECL version_attr, NULL
899 { "i", 0, 3, 0, 0, 0, 0, 1, "italic text style",
900 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
902 { "iframe", 0, 0, 0, 0, 0, 1, 2, "inline subwindow ",
903 DECL html_flow, NULL, NULL, DECL iframe_attrs, NULL
905 { "img", 0, 2, 2, 1, 0, 0, 1, "embedded image ",
906 EMPTY, NULL, DECL img_attrs, DECL align_attr, DECL src_alt_attrs
908 { "input", 0, 2, 2, 1, 0, 0, 1, "form control ",
909 EMPTY, NULL, DECL input_attrs , DECL align_attr, NULL
911 { "ins", 0, 0, 0, 0, 0, 0, 2, "inserted text",
912 DECL html_flow, NULL, DECL edit_attrs, NULL, NULL
914 { "isindex", 0, 2, 2, 1, 1, 1, 0, "single line prompt ",
915 EMPTY, NULL, NULL, DECL prompt_attrs, NULL
917 { "kbd", 0, 0, 0, 0, 0, 0, 1, "text to be entered by the user",
918 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
920 { "label", 0, 0, 0, 0, 0, 0, 1, "form field label text ",
921 DECL html_inline MODIFIER, NULL, DECL label_attrs , NULL, NULL
923 { "legend", 0, 0, 0, 0, 0, 0, 0, "fieldset legend ",
924 DECL html_inline, NULL, DECL legend_attrs , DECL align_attr, NULL
926 { "li", 0, 1, 1, 0, 0, 0, 0, "list item ",
927 DECL html_flow, NULL, DECL html_attrs, NULL, NULL
929 { "link", 0, 2, 2, 1, 0, 0, 0, "a media-independent link ",
930 EMPTY, NULL, DECL link_attrs, DECL target_attr, NULL
932 { "map", 0, 0, 0, 0, 0, 0, 2, "client-side image map ",
933 DECL map_contents , NULL, DECL html_attrs , NULL, DECL name_attr
935 { "menu", 0, 0, 0, 0, 1, 1, 0, "menu list ",
936 DECL blockli_elt , NULL, NULL, DECL compact_attrs, NULL
938 { "meta", 0, 2, 2, 1, 0, 0, 0, "generic metainformation ",
939 EMPTY, NULL, DECL meta_attrs , NULL , DECL content_attr
941 { "noframes", 0, 0, 0, 0, 0, 2, 0, "alternate content container for non frame-based rendering ",
942 DECL noframes_content, "body" , DECL html_attrs, NULL, NULL
944 { "noscript", 0, 0, 0, 0, 0, 0, 0, "alternate content container for non script-based rendering ",
945 DECL html_flow, "div", DECL html_attrs, NULL, NULL
947 { "object", 0, 0, 0, 0, 0, 0, 2, "generic embedded object ",
948 DECL object_contents , "div" , DECL object_attrs, DECL object_depr, NULL
950 { "ol", 0, 0, 0, 0, 0, 0, 0, "ordered list ",
951 DECL li_elt , "li" , DECL html_attrs, DECL ol_attrs, NULL
953 { "optgroup", 0, 0, 0, 0, 0, 0, 0, "option group ",
954 DECL option_elt , "option", DECL optgroup_attrs, NULL, DECL label_attr
956 { "option", 0, 1, 0, 0, 0, 0, 0, "selectable choice " ,
957 DECL html_pcdata, NULL, DECL option_attrs, NULL, NULL
959 { "p", 0, 1, 0, 0, 0, 0, 0, "paragraph ",
960 DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
962 { "param", 0, 2, 2, 1, 0, 0, 0, "named property value ",
963 EMPTY, NULL, DECL param_attrs, NULL, DECL name_attr
965 { "pre", 0, 0, 0, 0, 0, 0, 0, "preformatted text ",
966 DECL pre_content, NULL, DECL html_attrs, DECL width_attr, NULL
968 { "q", 0, 0, 0, 0, 0, 0, 1, "short inline quotation ",
969 DECL html_inline, NULL, DECL quote_attrs, NULL, NULL
971 { "s", 0, 3, 0, 0, 1, 1, 1, "strike-through text style",
972 DECL html_inline, NULL, NULL, DECL html_attrs, NULL
974 { "samp", 0, 0, 0, 0, 0, 0, 1, "sample program output, scripts, etc.",
975 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
977 { "script", 0, 0, 0, 0, 0, 0, 2, "script statements ",
978 DECL html_cdata, NULL, DECL script_attrs, DECL language_attr, DECL type_attr
980 { "select", 0, 0, 0, 0, 0, 0, 1, "option selector ",
981 DECL select_content, NULL, DECL select_attrs, NULL, NULL
983 { "small", 0, 3, 0, 0, 0, 0, 1, "small text style",
984 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
986 { "span", 0, 0, 0, 0, 0, 0, 1, "generic language/style container ",
987 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
989 { "strike", 0, 3, 0, 0, 1, 1, 1, "strike-through text",
990 DECL html_inline, NULL, NULL, DECL html_attrs, NULL
992 { "strong", 0, 3, 0, 0, 0, 0, 1, "strong emphasis",
993 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
995 { "style", 0, 0, 0, 0, 0, 0, 0, "style info ",
996 DECL html_cdata, NULL, DECL style_attrs, NULL, DECL type_attr
998 { "sub", 0, 3, 0, 0, 0, 0, 1, "subscript",
999 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
1001 { "sup", 0, 3, 0, 0, 0, 0, 1, "superscript ",
1002 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
1004 { "table", 0, 0, 0, 0, 0, 0, 0, "",
1005 DECL table_contents , "tr" , DECL table_attrs , DECL table_depr, NULL
1007 { "tbody", 1, 0, 0, 0, 0, 0, 0, "table body ",
1008 DECL tr_elt , "tr" , DECL talign_attrs, NULL, NULL
1010 { "td", 0, 0, 0, 0, 0, 0, 0, "table data cell",
1011 DECL html_flow, NULL, DECL th_td_attr, DECL th_td_depr, NULL
1013 { "textarea", 0, 0, 0, 0, 0, 0, 1, "multi-line text field ",
1014 DECL html_pcdata, NULL, DECL textarea_attrs, NULL, DECL rows_cols_attr
1016 { "tfoot", 0, 1, 0, 0, 0, 0, 0, "table footer ",
1017 DECL tr_elt , "tr" , DECL talign_attrs, NULL, NULL
1019 { "th", 0, 1, 0, 0, 0, 0, 0, "table header cell",
1020 DECL html_flow, NULL, DECL th_td_attr, DECL th_td_depr, NULL
1022 { "thead", 0, 1, 0, 0, 0, 0, 0, "table header ",
1023 DECL tr_elt , "tr" , DECL talign_attrs, NULL, NULL
1025 { "title", 0, 0, 0, 0, 0, 0, 0, "document title ",
1026 DECL html_pcdata, NULL, DECL i18n_attrs, NULL, NULL
1028 { "tr", 0, 0, 0, 0, 0, 0, 0, "table row ",
1029 DECL tr_contents , "td" , DECL talign_attrs, DECL bgcolor_attr, NULL
1031 { "tt", 0, 3, 0, 0, 0, 0, 1, "teletype or monospaced text style",
1032 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
1034 { "u", 0, 3, 0, 0, 1, 1, 1, "underlined text style",
1035 DECL html_inline, NULL, NULL, DECL html_attrs, NULL
1037 { "ul", 0, 0, 0, 0, 0, 0, 0, "unordered list ",
1038 DECL li_elt , "li" , DECL html_attrs, DECL ul_depr, NULL
1040 { "var", 0, 0, 0, 0, 0, 0, 1, "instance of a variable or program argument",
1041 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
1046 * start tags that imply the end of current element
1048 static const char * const htmlStartClose[] = {
1049 "form", "form", "p", "hr", "h1", "h2", "h3", "h4", "h5", "h6",
1050 "dl", "ul", "ol", "menu", "dir", "address", "pre",
1051 "listing", "xmp", "head", NULL,
1054 "body", "head", "style", "link", "title", "p", NULL,
1055 "frameset", "head", "style", "link", "title", "p", NULL,
1056 "li", "p", "h1", "h2", "h3", "h4", "h5", "h6", "dl", "address",
1057 "pre", "listing", "xmp", "head", "li", NULL,
1058 "hr", "p", "head", NULL,
1059 "h1", "p", "head", NULL,
1060 "h2", "p", "head", NULL,
1061 "h3", "p", "head", NULL,
1062 "h4", "p", "head", NULL,
1063 "h5", "p", "head", NULL,
1064 "h6", "p", "head", NULL,
1065 "dir", "p", "head", NULL,
1066 "address", "p", "head", "ul", NULL,
1067 "pre", "p", "head", "ul", NULL,
1068 "listing", "p", "head", NULL,
1069 "xmp", "p", "head", NULL,
1070 "blockquote", "p", "head", NULL,
1071 "dl", "p", "dt", "menu", "dir", "address", "pre", "listing",
1072 "xmp", "head", NULL,
1073 "dt", "p", "menu", "dir", "address", "pre", "listing", "xmp",
1075 "dd", "p", "menu", "dir", "address", "pre", "listing", "xmp",
1077 "ul", "p", "head", "ol", "menu", "dir", "address", "pre",
1078 "listing", "xmp", NULL,
1079 "ol", "p", "head", "ul", NULL,
1080 "menu", "p", "head", "ul", NULL,
1081 "p", "p", "head", "h1", "h2", "h3", "h4", "h5", "h6", FONTSTYLE, NULL,
1082 "div", "p", "head", NULL,
1083 "noscript", "p", "head", NULL,
1084 "center", "font", "b", "i", "p", "head", NULL,
1086 "caption", "p", NULL,
1087 "colgroup", "caption", "colgroup", "col", "p", NULL,
1088 "col", "caption", "col", "p", NULL,
1089 "table", "p", "head", "h1", "h2", "h3", "h4", "h5", "h6", "pre",
1090 "listing", "xmp", "a", NULL,
1091 "th", "th", "td", "p", "span", "font", "a", "b", "i", "u", NULL,
1092 "td", "th", "td", "p", "span", "font", "a", "b", "i", "u", NULL,
1093 "tr", "th", "td", "tr", "caption", "col", "colgroup", "p", NULL,
1094 "thead", "caption", "col", "colgroup", NULL,
1095 "tfoot", "th", "td", "tr", "caption", "col", "colgroup", "thead",
1097 "tbody", "th", "td", "tr", "caption", "col", "colgroup", "thead",
1098 "tfoot", "tbody", "p", NULL,
1099 "optgroup", "option", NULL,
1100 "option", "option", NULL,
1101 "fieldset", "legend", "p", "head", "h1", "h2", "h3", "h4", "h5", "h6",
1102 "pre", "listing", "xmp", "a", NULL,
1107 * The list of HTML elements which are supposed not to have
1108 * CDATA content and where a p element will be implied
1110 * TODO: extend that list by reading the HTML SGML DTD on
1113 static const char *const htmlNoContentElements[] = {
1120 * The list of HTML attributes which are of content %Script;
1121 * NOTE: when adding ones, check htmlIsScriptAttribute() since
1122 * it assumes the name starts with 'on'
1124 static const char *const htmlScriptAttributes[] = {
1146 * This table is used by the htmlparser to know what to do with
1147 * broken html pages. By assigning different priorities to different
1148 * elements the parser can decide how to handle extra endtags.
1149 * Endtags are only allowed to close elements with lower or equal
1158 static const elementPriority htmlEndPriority[] = {
1170 {NULL, 100} /* Default priority */
1173 static const char** htmlStartCloseIndex[100];
1174 static int htmlStartCloseIndexinitialized = 0;
1176 /************************************************************************
1178 * functions to handle HTML specific data *
1180 ************************************************************************/
1183 * htmlInitAutoClose:
1185 * Initialize the htmlStartCloseIndex for fast lookup of closing tags names.
1186 * This is not reentrant. Call xmlInitParser() once before processing in
1187 * case of use in multithreaded programs.
1190 htmlInitAutoClose(void) {
1193 if (htmlStartCloseIndexinitialized) return;
1195 for (indx = 0;indx < 100;indx ++) htmlStartCloseIndex[indx] = NULL;
1197 while ((htmlStartClose[i] != NULL) && (indx < 100 - 1)) {
1198 htmlStartCloseIndex[indx++] = (const char**) &htmlStartClose[i];
1199 while (htmlStartClose[i] != NULL) i++;
1202 htmlStartCloseIndexinitialized = 1;
1207 * @tag: The tag name in lowercase
1209 * Lookup the HTML tag in the ElementTable
1211 * Returns the related htmlElemDescPtr or NULL if not found.
1213 const htmlElemDesc *
1214 htmlTagLookup(const xmlChar *tag) {
1217 for (i = 0; i < (sizeof(html40ElementTable) /
1218 sizeof(html40ElementTable[0]));i++) {
1219 if (!xmlStrcasecmp(tag, BAD_CAST html40ElementTable[i].name))
1220 return((htmlElemDescPtr) &html40ElementTable[i]);
1226 * htmlGetEndPriority:
1227 * @name: The name of the element to look up the priority for.
1229 * Return value: The "endtag" priority.
1232 htmlGetEndPriority (const xmlChar *name) {
1235 while ((htmlEndPriority[i].name != NULL) &&
1236 (!xmlStrEqual((const xmlChar *)htmlEndPriority[i].name, name)))
1239 return(htmlEndPriority[i].priority);
1244 * htmlCheckAutoClose:
1245 * @newtag: The new tag name
1246 * @oldtag: The old tag name
1248 * Checks whether the new tag is one of the registered valid tags for
1250 * Initialize the htmlStartCloseIndex for fast lookup of closing tags names.
1252 * Returns 0 if no, 1 if yes.
1255 htmlCheckAutoClose(const xmlChar * newtag, const xmlChar * oldtag)
1258 const char **closed = NULL;
1260 if (htmlStartCloseIndexinitialized == 0)
1261 htmlInitAutoClose();
1263 /* inefficient, but not a big deal */
1264 for (indx = 0; indx < 100; indx++) {
1265 closed = htmlStartCloseIndex[indx];
1268 if (xmlStrEqual(BAD_CAST * closed, newtag))
1272 i = closed - htmlStartClose;
1274 while (htmlStartClose[i] != NULL) {
1275 if (xmlStrEqual(BAD_CAST htmlStartClose[i], oldtag)) {
1284 * htmlAutoCloseOnClose:
1285 * @ctxt: an HTML parser context
1286 * @newtag: The new tag name
1287 * @force: force the tag closure
1289 * The HTML DTD allows an ending tag to implicitly close other tags.
1292 htmlAutoCloseOnClose(htmlParserCtxtPtr ctxt, const xmlChar * newtag)
1294 const htmlElemDesc *info;
1297 priority = htmlGetEndPriority(newtag);
1299 for (i = (ctxt->nameNr - 1); i >= 0; i--) {
1301 if (xmlStrEqual(newtag, ctxt->nameTab[i]))
1304 * A missplaced endtag can only close elements with lower
1305 * or equal priority, so if we find an element with higher
1306 * priority before we find an element with
1307 * matching name, we just ignore this endtag
1309 if (htmlGetEndPriority(ctxt->nameTab[i]) > priority)
1315 while (!xmlStrEqual(newtag, ctxt->name)) {
1316 info = htmlTagLookup(ctxt->name);
1317 if ((info != NULL) && (info->endTag == 3)) {
1318 htmlParseErr(ctxt, XML_ERR_TAG_NAME_MISMATCH,
1319 "Opening and ending tag mismatch: %s and %s\n",
1320 newtag, ctxt->name);
1322 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
1323 ctxt->sax->endElement(ctxt->userData, ctxt->name);
1329 * htmlAutoCloseOnEnd:
1330 * @ctxt: an HTML parser context
1332 * Close all remaining tags at the end of the stream
1335 htmlAutoCloseOnEnd(htmlParserCtxtPtr ctxt)
1339 if (ctxt->nameNr == 0)
1341 for (i = (ctxt->nameNr - 1); i >= 0; i--) {
1342 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
1343 ctxt->sax->endElement(ctxt->userData, ctxt->name);
1350 * @ctxt: an HTML parser context
1351 * @newtag: The new tag name or NULL
1353 * The HTML DTD allows a tag to implicitly close other tags.
1354 * The list is kept in htmlStartClose array. This function is
1355 * called when a new tag has been detected and generates the
1356 * appropriates closes if possible/needed.
1357 * If newtag is NULL this mean we are at the end of the resource
1358 * and we should check
1361 htmlAutoClose(htmlParserCtxtPtr ctxt, const xmlChar * newtag)
1363 while ((newtag != NULL) && (ctxt->name != NULL) &&
1364 (htmlCheckAutoClose(newtag, ctxt->name))) {
1365 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
1366 ctxt->sax->endElement(ctxt->userData, ctxt->name);
1369 if (newtag == NULL) {
1370 htmlAutoCloseOnEnd(ctxt);
1373 while ((newtag == NULL) && (ctxt->name != NULL) &&
1374 ((xmlStrEqual(ctxt->name, BAD_CAST "head")) ||
1375 (xmlStrEqual(ctxt->name, BAD_CAST "body")) ||
1376 (xmlStrEqual(ctxt->name, BAD_CAST "html")))) {
1377 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
1378 ctxt->sax->endElement(ctxt->userData, ctxt->name);
1385 * @doc: the HTML document
1386 * @name: The tag name
1387 * @elem: the HTML element
1389 * The HTML DTD allows a tag to implicitly close other tags.
1390 * The list is kept in htmlStartClose array. This function checks
1391 * if the element or one of it's children would autoclose the
1394 * Returns 1 if autoclose, 0 otherwise
1397 htmlAutoCloseTag(htmlDocPtr doc, const xmlChar *name, htmlNodePtr elem) {
1400 if (elem == NULL) return(1);
1401 if (xmlStrEqual(name, elem->name)) return(0);
1402 if (htmlCheckAutoClose(elem->name, name)) return(1);
1403 child = elem->children;
1404 while (child != NULL) {
1405 if (htmlAutoCloseTag(doc, name, child)) return(1);
1406 child = child->next;
1413 * @doc: the HTML document
1414 * @elem: the HTML element
1416 * The HTML DTD allows a tag to implicitly close other tags.
1417 * The list is kept in htmlStartClose array. This function checks
1418 * if a tag is autoclosed by one of it's child
1420 * Returns 1 if autoclosed, 0 otherwise
1423 htmlIsAutoClosed(htmlDocPtr doc, htmlNodePtr elem) {
1426 if (elem == NULL) return(1);
1427 child = elem->children;
1428 while (child != NULL) {
1429 if (htmlAutoCloseTag(doc, elem->name, child)) return(1);
1430 child = child->next;
1437 * @ctxt: an HTML parser context
1438 * @newtag: The new tag name
1440 * The HTML DTD allows a tag to exists only implicitly
1441 * called when a new tag has been detected and generates the
1442 * appropriates implicit tags if missing
1445 htmlCheckImplied(htmlParserCtxtPtr ctxt, const xmlChar *newtag) {
1448 if (ctxt->options & HTML_PARSE_NOIMPLIED)
1450 if (!htmlOmittedDefaultValue)
1452 if (xmlStrEqual(newtag, BAD_CAST"html"))
1454 if (ctxt->nameNr <= 0) {
1455 htmlnamePush(ctxt, BAD_CAST"html");
1456 if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
1457 ctxt->sax->startElement(ctxt->userData, BAD_CAST"html", NULL);
1459 if ((xmlStrEqual(newtag, BAD_CAST"body")) || (xmlStrEqual(newtag, BAD_CAST"head")))
1461 if ((ctxt->nameNr <= 1) &&
1462 ((xmlStrEqual(newtag, BAD_CAST"script")) ||
1463 (xmlStrEqual(newtag, BAD_CAST"style")) ||
1464 (xmlStrEqual(newtag, BAD_CAST"meta")) ||
1465 (xmlStrEqual(newtag, BAD_CAST"link")) ||
1466 (xmlStrEqual(newtag, BAD_CAST"title")) ||
1467 (xmlStrEqual(newtag, BAD_CAST"base")))) {
1468 if (ctxt->html >= 3) {
1469 /* we already saw or generated an <head> before */
1473 * dropped OBJECT ... i you put it first BODY will be
1476 htmlnamePush(ctxt, BAD_CAST"head");
1477 if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
1478 ctxt->sax->startElement(ctxt->userData, BAD_CAST"head", NULL);
1479 } else if ((!xmlStrEqual(newtag, BAD_CAST"noframes")) &&
1480 (!xmlStrEqual(newtag, BAD_CAST"frame")) &&
1481 (!xmlStrEqual(newtag, BAD_CAST"frameset"))) {
1482 if (ctxt->html >= 10) {
1483 /* we already saw or generated a <body> before */
1486 for (i = 0;i < ctxt->nameNr;i++) {
1487 if (xmlStrEqual(ctxt->nameTab[i], BAD_CAST"body")) {
1490 if (xmlStrEqual(ctxt->nameTab[i], BAD_CAST"head")) {
1495 htmlnamePush(ctxt, BAD_CAST"body");
1496 if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
1497 ctxt->sax->startElement(ctxt->userData, BAD_CAST"body", NULL);
1502 * htmlCheckParagraph
1503 * @ctxt: an HTML parser context
1505 * Check whether a p element need to be implied before inserting
1506 * characters in the current element.
1508 * Returns 1 if a paragraph has been inserted, 0 if not and -1
1513 htmlCheckParagraph(htmlParserCtxtPtr ctxt) {
1521 htmlAutoClose(ctxt, BAD_CAST"p");
1522 htmlCheckImplied(ctxt, BAD_CAST"p");
1523 htmlnamePush(ctxt, BAD_CAST"p");
1524 if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
1525 ctxt->sax->startElement(ctxt->userData, BAD_CAST"p", NULL);
1528 if (!htmlOmittedDefaultValue)
1530 for (i = 0; htmlNoContentElements[i] != NULL; i++) {
1531 if (xmlStrEqual(tag, BAD_CAST htmlNoContentElements[i])) {
1532 htmlAutoClose(ctxt, BAD_CAST"p");
1533 htmlCheckImplied(ctxt, BAD_CAST"p");
1534 htmlnamePush(ctxt, BAD_CAST"p");
1535 if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
1536 ctxt->sax->startElement(ctxt->userData, BAD_CAST"p", NULL);
1544 * htmlIsScriptAttribute:
1545 * @name: an attribute name
1547 * Check if an attribute is of content type Script
1549 * Returns 1 is the attribute is a script 0 otherwise
1552 htmlIsScriptAttribute(const xmlChar *name) {
1558 * all script attributes start with 'on'
1560 if ((name[0] != 'o') || (name[1] != 'n'))
1563 i < sizeof(htmlScriptAttributes)/sizeof(htmlScriptAttributes[0]);
1565 if (xmlStrEqual(name, (const xmlChar *) htmlScriptAttributes[i]))
1571 /************************************************************************
1573 * The list of HTML predefined entities *
1575 ************************************************************************/
1578 static const htmlEntityDesc html40EntitiesTable[] = {
1580 * the 4 absolute ones, plus apostrophe.
1582 { 34, "quot", "quotation mark = APL quote, U+0022 ISOnum" },
1583 { 38, "amp", "ampersand, U+0026 ISOnum" },
1584 { 39, "apos", "single quote" },
1585 { 60, "lt", "less-than sign, U+003C ISOnum" },
1586 { 62, "gt", "greater-than sign, U+003E ISOnum" },
1589 * A bunch still in the 128-255 range
1590 * Replacing them depend really on the charset used.
1592 { 160, "nbsp", "no-break space = non-breaking space, U+00A0 ISOnum" },
1593 { 161, "iexcl","inverted exclamation mark, U+00A1 ISOnum" },
1594 { 162, "cent", "cent sign, U+00A2 ISOnum" },
1595 { 163, "pound","pound sign, U+00A3 ISOnum" },
1596 { 164, "curren","currency sign, U+00A4 ISOnum" },
1597 { 165, "yen", "yen sign = yuan sign, U+00A5 ISOnum" },
1598 { 166, "brvbar","broken bar = broken vertical bar, U+00A6 ISOnum" },
1599 { 167, "sect", "section sign, U+00A7 ISOnum" },
1600 { 168, "uml", "diaeresis = spacing diaeresis, U+00A8 ISOdia" },
1601 { 169, "copy", "copyright sign, U+00A9 ISOnum" },
1602 { 170, "ordf", "feminine ordinal indicator, U+00AA ISOnum" },
1603 { 171, "laquo","left-pointing double angle quotation mark = left pointing guillemet, U+00AB ISOnum" },
1604 { 172, "not", "not sign, U+00AC ISOnum" },
1605 { 173, "shy", "soft hyphen = discretionary hyphen, U+00AD ISOnum" },
1606 { 174, "reg", "registered sign = registered trade mark sign, U+00AE ISOnum" },
1607 { 175, "macr", "macron = spacing macron = overline = APL overbar, U+00AF ISOdia" },
1608 { 176, "deg", "degree sign, U+00B0 ISOnum" },
1609 { 177, "plusmn","plus-minus sign = plus-or-minus sign, U+00B1 ISOnum" },
1610 { 178, "sup2", "superscript two = superscript digit two = squared, U+00B2 ISOnum" },
1611 { 179, "sup3", "superscript three = superscript digit three = cubed, U+00B3 ISOnum" },
1612 { 180, "acute","acute accent = spacing acute, U+00B4 ISOdia" },
1613 { 181, "micro","micro sign, U+00B5 ISOnum" },
1614 { 182, "para", "pilcrow sign = paragraph sign, U+00B6 ISOnum" },
1615 { 183, "middot","middle dot = Georgian comma Greek middle dot, U+00B7 ISOnum" },
1616 { 184, "cedil","cedilla = spacing cedilla, U+00B8 ISOdia" },
1617 { 185, "sup1", "superscript one = superscript digit one, U+00B9 ISOnum" },
1618 { 186, "ordm", "masculine ordinal indicator, U+00BA ISOnum" },
1619 { 187, "raquo","right-pointing double angle quotation mark right pointing guillemet, U+00BB ISOnum" },
1620 { 188, "frac14","vulgar fraction one quarter = fraction one quarter, U+00BC ISOnum" },
1621 { 189, "frac12","vulgar fraction one half = fraction one half, U+00BD ISOnum" },
1622 { 190, "frac34","vulgar fraction three quarters = fraction three quarters, U+00BE ISOnum" },
1623 { 191, "iquest","inverted question mark = turned question mark, U+00BF ISOnum" },
1624 { 192, "Agrave","latin capital letter A with grave = latin capital letter A grave, U+00C0 ISOlat1" },
1625 { 193, "Aacute","latin capital letter A with acute, U+00C1 ISOlat1" },
1626 { 194, "Acirc","latin capital letter A with circumflex, U+00C2 ISOlat1" },
1627 { 195, "Atilde","latin capital letter A with tilde, U+00C3 ISOlat1" },
1628 { 196, "Auml", "latin capital letter A with diaeresis, U+00C4 ISOlat1" },
1629 { 197, "Aring","latin capital letter A with ring above = latin capital letter A ring, U+00C5 ISOlat1" },
1630 { 198, "AElig","latin capital letter AE = latin capital ligature AE, U+00C6 ISOlat1" },
1631 { 199, "Ccedil","latin capital letter C with cedilla, U+00C7 ISOlat1" },
1632 { 200, "Egrave","latin capital letter E with grave, U+00C8 ISOlat1" },
1633 { 201, "Eacute","latin capital letter E with acute, U+00C9 ISOlat1" },
1634 { 202, "Ecirc","latin capital letter E with circumflex, U+00CA ISOlat1" },
1635 { 203, "Euml", "latin capital letter E with diaeresis, U+00CB ISOlat1" },
1636 { 204, "Igrave","latin capital letter I with grave, U+00CC ISOlat1" },
1637 { 205, "Iacute","latin capital letter I with acute, U+00CD ISOlat1" },
1638 { 206, "Icirc","latin capital letter I with circumflex, U+00CE ISOlat1" },
1639 { 207, "Iuml", "latin capital letter I with diaeresis, U+00CF ISOlat1" },
1640 { 208, "ETH", "latin capital letter ETH, U+00D0 ISOlat1" },
1641 { 209, "Ntilde","latin capital letter N with tilde, U+00D1 ISOlat1" },
1642 { 210, "Ograve","latin capital letter O with grave, U+00D2 ISOlat1" },
1643 { 211, "Oacute","latin capital letter O with acute, U+00D3 ISOlat1" },
1644 { 212, "Ocirc","latin capital letter O with circumflex, U+00D4 ISOlat1" },
1645 { 213, "Otilde","latin capital letter O with tilde, U+00D5 ISOlat1" },
1646 { 214, "Ouml", "latin capital letter O with diaeresis, U+00D6 ISOlat1" },
1647 { 215, "times","multiplication sign, U+00D7 ISOnum" },
1648 { 216, "Oslash","latin capital letter O with stroke latin capital letter O slash, U+00D8 ISOlat1" },
1649 { 217, "Ugrave","latin capital letter U with grave, U+00D9 ISOlat1" },
1650 { 218, "Uacute","latin capital letter U with acute, U+00DA ISOlat1" },
1651 { 219, "Ucirc","latin capital letter U with circumflex, U+00DB ISOlat1" },
1652 { 220, "Uuml", "latin capital letter U with diaeresis, U+00DC ISOlat1" },
1653 { 221, "Yacute","latin capital letter Y with acute, U+00DD ISOlat1" },
1654 { 222, "THORN","latin capital letter THORN, U+00DE ISOlat1" },
1655 { 223, "szlig","latin small letter sharp s = ess-zed, U+00DF ISOlat1" },
1656 { 224, "agrave","latin small letter a with grave = latin small letter a grave, U+00E0 ISOlat1" },
1657 { 225, "aacute","latin small letter a with acute, U+00E1 ISOlat1" },
1658 { 226, "acirc","latin small letter a with circumflex, U+00E2 ISOlat1" },
1659 { 227, "atilde","latin small letter a with tilde, U+00E3 ISOlat1" },
1660 { 228, "auml", "latin small letter a with diaeresis, U+00E4 ISOlat1" },
1661 { 229, "aring","latin small letter a with ring above = latin small letter a ring, U+00E5 ISOlat1" },
1662 { 230, "aelig","latin small letter ae = latin small ligature ae, U+00E6 ISOlat1" },
1663 { 231, "ccedil","latin small letter c with cedilla, U+00E7 ISOlat1" },
1664 { 232, "egrave","latin small letter e with grave, U+00E8 ISOlat1" },
1665 { 233, "eacute","latin small letter e with acute, U+00E9 ISOlat1" },
1666 { 234, "ecirc","latin small letter e with circumflex, U+00EA ISOlat1" },
1667 { 235, "euml", "latin small letter e with diaeresis, U+00EB ISOlat1" },
1668 { 236, "igrave","latin small letter i with grave, U+00EC ISOlat1" },
1669 { 237, "iacute","latin small letter i with acute, U+00ED ISOlat1" },
1670 { 238, "icirc","latin small letter i with circumflex, U+00EE ISOlat1" },
1671 { 239, "iuml", "latin small letter i with diaeresis, U+00EF ISOlat1" },
1672 { 240, "eth", "latin small letter eth, U+00F0 ISOlat1" },
1673 { 241, "ntilde","latin small letter n with tilde, U+00F1 ISOlat1" },
1674 { 242, "ograve","latin small letter o with grave, U+00F2 ISOlat1" },
1675 { 243, "oacute","latin small letter o with acute, U+00F3 ISOlat1" },
1676 { 244, "ocirc","latin small letter o with circumflex, U+00F4 ISOlat1" },
1677 { 245, "otilde","latin small letter o with tilde, U+00F5 ISOlat1" },
1678 { 246, "ouml", "latin small letter o with diaeresis, U+00F6 ISOlat1" },
1679 { 247, "divide","division sign, U+00F7 ISOnum" },
1680 { 248, "oslash","latin small letter o with stroke, = latin small letter o slash, U+00F8 ISOlat1" },
1681 { 249, "ugrave","latin small letter u with grave, U+00F9 ISOlat1" },
1682 { 250, "uacute","latin small letter u with acute, U+00FA ISOlat1" },
1683 { 251, "ucirc","latin small letter u with circumflex, U+00FB ISOlat1" },
1684 { 252, "uuml", "latin small letter u with diaeresis, U+00FC ISOlat1" },
1685 { 253, "yacute","latin small letter y with acute, U+00FD ISOlat1" },
1686 { 254, "thorn","latin small letter thorn with, U+00FE ISOlat1" },
1687 { 255, "yuml", "latin small letter y with diaeresis, U+00FF ISOlat1" },
1689 { 338, "OElig","latin capital ligature OE, U+0152 ISOlat2" },
1690 { 339, "oelig","latin small ligature oe, U+0153 ISOlat2" },
1691 { 352, "Scaron","latin capital letter S with caron, U+0160 ISOlat2" },
1692 { 353, "scaron","latin small letter s with caron, U+0161 ISOlat2" },
1693 { 376, "Yuml", "latin capital letter Y with diaeresis, U+0178 ISOlat2" },
1696 * Anything below should really be kept as entities references
1698 { 402, "fnof", "latin small f with hook = function = florin, U+0192 ISOtech" },
1700 { 710, "circ", "modifier letter circumflex accent, U+02C6 ISOpub" },
1701 { 732, "tilde","small tilde, U+02DC ISOdia" },
1703 { 913, "Alpha","greek capital letter alpha, U+0391" },
1704 { 914, "Beta", "greek capital letter beta, U+0392" },
1705 { 915, "Gamma","greek capital letter gamma, U+0393 ISOgrk3" },
1706 { 916, "Delta","greek capital letter delta, U+0394 ISOgrk3" },
1707 { 917, "Epsilon","greek capital letter epsilon, U+0395" },
1708 { 918, "Zeta", "greek capital letter zeta, U+0396" },
1709 { 919, "Eta", "greek capital letter eta, U+0397" },
1710 { 920, "Theta","greek capital letter theta, U+0398 ISOgrk3" },
1711 { 921, "Iota", "greek capital letter iota, U+0399" },
1712 { 922, "Kappa","greek capital letter kappa, U+039A" },
1713 { 923, "Lambda", "greek capital letter lambda, U+039B ISOgrk3" },
1714 { 924, "Mu", "greek capital letter mu, U+039C" },
1715 { 925, "Nu", "greek capital letter nu, U+039D" },
1716 { 926, "Xi", "greek capital letter xi, U+039E ISOgrk3" },
1717 { 927, "Omicron","greek capital letter omicron, U+039F" },
1718 { 928, "Pi", "greek capital letter pi, U+03A0 ISOgrk3" },
1719 { 929, "Rho", "greek capital letter rho, U+03A1" },
1720 { 931, "Sigma","greek capital letter sigma, U+03A3 ISOgrk3" },
1721 { 932, "Tau", "greek capital letter tau, U+03A4" },
1722 { 933, "Upsilon","greek capital letter upsilon, U+03A5 ISOgrk3" },
1723 { 934, "Phi", "greek capital letter phi, U+03A6 ISOgrk3" },
1724 { 935, "Chi", "greek capital letter chi, U+03A7" },
1725 { 936, "Psi", "greek capital letter psi, U+03A8 ISOgrk3" },
1726 { 937, "Omega","greek capital letter omega, U+03A9 ISOgrk3" },
1728 { 945, "alpha","greek small letter alpha, U+03B1 ISOgrk3" },
1729 { 946, "beta", "greek small letter beta, U+03B2 ISOgrk3" },
1730 { 947, "gamma","greek small letter gamma, U+03B3 ISOgrk3" },
1731 { 948, "delta","greek small letter delta, U+03B4 ISOgrk3" },
1732 { 949, "epsilon","greek small letter epsilon, U+03B5 ISOgrk3" },
1733 { 950, "zeta", "greek small letter zeta, U+03B6 ISOgrk3" },
1734 { 951, "eta", "greek small letter eta, U+03B7 ISOgrk3" },
1735 { 952, "theta","greek small letter theta, U+03B8 ISOgrk3" },
1736 { 953, "iota", "greek small letter iota, U+03B9 ISOgrk3" },
1737 { 954, "kappa","greek small letter kappa, U+03BA ISOgrk3" },
1738 { 955, "lambda","greek small letter lambda, U+03BB ISOgrk3" },
1739 { 956, "mu", "greek small letter mu, U+03BC ISOgrk3" },
1740 { 957, "nu", "greek small letter nu, U+03BD ISOgrk3" },
1741 { 958, "xi", "greek small letter xi, U+03BE ISOgrk3" },
1742 { 959, "omicron","greek small letter omicron, U+03BF NEW" },
1743 { 960, "pi", "greek small letter pi, U+03C0 ISOgrk3" },
1744 { 961, "rho", "greek small letter rho, U+03C1 ISOgrk3" },
1745 { 962, "sigmaf","greek small letter final sigma, U+03C2 ISOgrk3" },
1746 { 963, "sigma","greek small letter sigma, U+03C3 ISOgrk3" },
1747 { 964, "tau", "greek small letter tau, U+03C4 ISOgrk3" },
1748 { 965, "upsilon","greek small letter upsilon, U+03C5 ISOgrk3" },
1749 { 966, "phi", "greek small letter phi, U+03C6 ISOgrk3" },
1750 { 967, "chi", "greek small letter chi, U+03C7 ISOgrk3" },
1751 { 968, "psi", "greek small letter psi, U+03C8 ISOgrk3" },
1752 { 969, "omega","greek small letter omega, U+03C9 ISOgrk3" },
1753 { 977, "thetasym","greek small letter theta symbol, U+03D1 NEW" },
1754 { 978, "upsih","greek upsilon with hook symbol, U+03D2 NEW" },
1755 { 982, "piv", "greek pi symbol, U+03D6 ISOgrk3" },
1757 { 8194, "ensp", "en space, U+2002 ISOpub" },
1758 { 8195, "emsp", "em space, U+2003 ISOpub" },
1759 { 8201, "thinsp","thin space, U+2009 ISOpub" },
1760 { 8204, "zwnj", "zero width non-joiner, U+200C NEW RFC 2070" },
1761 { 8205, "zwj", "zero width joiner, U+200D NEW RFC 2070" },
1762 { 8206, "lrm", "left-to-right mark, U+200E NEW RFC 2070" },
1763 { 8207, "rlm", "right-to-left mark, U+200F NEW RFC 2070" },
1764 { 8211, "ndash","en dash, U+2013 ISOpub" },
1765 { 8212, "mdash","em dash, U+2014 ISOpub" },
1766 { 8216, "lsquo","left single quotation mark, U+2018 ISOnum" },
1767 { 8217, "rsquo","right single quotation mark, U+2019 ISOnum" },
1768 { 8218, "sbquo","single low-9 quotation mark, U+201A NEW" },
1769 { 8220, "ldquo","left double quotation mark, U+201C ISOnum" },
1770 { 8221, "rdquo","right double quotation mark, U+201D ISOnum" },
1771 { 8222, "bdquo","double low-9 quotation mark, U+201E NEW" },
1772 { 8224, "dagger","dagger, U+2020 ISOpub" },
1773 { 8225, "Dagger","double dagger, U+2021 ISOpub" },
1775 { 8226, "bull", "bullet = black small circle, U+2022 ISOpub" },
1776 { 8230, "hellip","horizontal ellipsis = three dot leader, U+2026 ISOpub" },
1778 { 8240, "permil","per mille sign, U+2030 ISOtech" },
1780 { 8242, "prime","prime = minutes = feet, U+2032 ISOtech" },
1781 { 8243, "Prime","double prime = seconds = inches, U+2033 ISOtech" },
1783 { 8249, "lsaquo","single left-pointing angle quotation mark, U+2039 ISO proposed" },
1784 { 8250, "rsaquo","single right-pointing angle quotation mark, U+203A ISO proposed" },
1786 { 8254, "oline","overline = spacing overscore, U+203E NEW" },
1787 { 8260, "frasl","fraction slash, U+2044 NEW" },
1789 { 8364, "euro", "euro sign, U+20AC NEW" },
1791 { 8465, "image","blackletter capital I = imaginary part, U+2111 ISOamso" },
1792 { 8472, "weierp","script capital P = power set = Weierstrass p, U+2118 ISOamso" },
1793 { 8476, "real", "blackletter capital R = real part symbol, U+211C ISOamso" },
1794 { 8482, "trade","trade mark sign, U+2122 ISOnum" },
1795 { 8501, "alefsym","alef symbol = first transfinite cardinal, U+2135 NEW" },
1796 { 8592, "larr", "leftwards arrow, U+2190 ISOnum" },
1797 { 8593, "uarr", "upwards arrow, U+2191 ISOnum" },
1798 { 8594, "rarr", "rightwards arrow, U+2192 ISOnum" },
1799 { 8595, "darr", "downwards arrow, U+2193 ISOnum" },
1800 { 8596, "harr", "left right arrow, U+2194 ISOamsa" },
1801 { 8629, "crarr","downwards arrow with corner leftwards = carriage return, U+21B5 NEW" },
1802 { 8656, "lArr", "leftwards double arrow, U+21D0 ISOtech" },
1803 { 8657, "uArr", "upwards double arrow, U+21D1 ISOamsa" },
1804 { 8658, "rArr", "rightwards double arrow, U+21D2 ISOtech" },
1805 { 8659, "dArr", "downwards double arrow, U+21D3 ISOamsa" },
1806 { 8660, "hArr", "left right double arrow, U+21D4 ISOamsa" },
1808 { 8704, "forall","for all, U+2200 ISOtech" },
1809 { 8706, "part", "partial differential, U+2202 ISOtech" },
1810 { 8707, "exist","there exists, U+2203 ISOtech" },
1811 { 8709, "empty","empty set = null set = diameter, U+2205 ISOamso" },
1812 { 8711, "nabla","nabla = backward difference, U+2207 ISOtech" },
1813 { 8712, "isin", "element of, U+2208 ISOtech" },
1814 { 8713, "notin","not an element of, U+2209 ISOtech" },
1815 { 8715, "ni", "contains as member, U+220B ISOtech" },
1816 { 8719, "prod", "n-ary product = product sign, U+220F ISOamsb" },
1817 { 8721, "sum", "n-ary summation, U+2211 ISOamsb" },
1818 { 8722, "minus","minus sign, U+2212 ISOtech" },
1819 { 8727, "lowast","asterisk operator, U+2217 ISOtech" },
1820 { 8730, "radic","square root = radical sign, U+221A ISOtech" },
1821 { 8733, "prop", "proportional to, U+221D ISOtech" },
1822 { 8734, "infin","infinity, U+221E ISOtech" },
1823 { 8736, "ang", "angle, U+2220 ISOamso" },
1824 { 8743, "and", "logical and = wedge, U+2227 ISOtech" },
1825 { 8744, "or", "logical or = vee, U+2228 ISOtech" },
1826 { 8745, "cap", "intersection = cap, U+2229 ISOtech" },
1827 { 8746, "cup", "union = cup, U+222A ISOtech" },
1828 { 8747, "int", "integral, U+222B ISOtech" },
1829 { 8756, "there4","therefore, U+2234 ISOtech" },
1830 { 8764, "sim", "tilde operator = varies with = similar to, U+223C ISOtech" },
1831 { 8773, "cong", "approximately equal to, U+2245 ISOtech" },
1832 { 8776, "asymp","almost equal to = asymptotic to, U+2248 ISOamsr" },
1833 { 8800, "ne", "not equal to, U+2260 ISOtech" },
1834 { 8801, "equiv","identical to, U+2261 ISOtech" },
1835 { 8804, "le", "less-than or equal to, U+2264 ISOtech" },
1836 { 8805, "ge", "greater-than or equal to, U+2265 ISOtech" },
1837 { 8834, "sub", "subset of, U+2282 ISOtech" },
1838 { 8835, "sup", "superset of, U+2283 ISOtech" },
1839 { 8836, "nsub", "not a subset of, U+2284 ISOamsn" },
1840 { 8838, "sube", "subset of or equal to, U+2286 ISOtech" },
1841 { 8839, "supe", "superset of or equal to, U+2287 ISOtech" },
1842 { 8853, "oplus","circled plus = direct sum, U+2295 ISOamsb" },
1843 { 8855, "otimes","circled times = vector product, U+2297 ISOamsb" },
1844 { 8869, "perp", "up tack = orthogonal to = perpendicular, U+22A5 ISOtech" },
1845 { 8901, "sdot", "dot operator, U+22C5 ISOamsb" },
1846 { 8968, "lceil","left ceiling = apl upstile, U+2308 ISOamsc" },
1847 { 8969, "rceil","right ceiling, U+2309 ISOamsc" },
1848 { 8970, "lfloor","left floor = apl downstile, U+230A ISOamsc" },
1849 { 8971, "rfloor","right floor, U+230B ISOamsc" },
1850 { 9001, "lang", "left-pointing angle bracket = bra, U+2329 ISOtech" },
1851 { 9002, "rang", "right-pointing angle bracket = ket, U+232A ISOtech" },
1852 { 9674, "loz", "lozenge, U+25CA ISOpub" },
1854 { 9824, "spades","black spade suit, U+2660 ISOpub" },
1855 { 9827, "clubs","black club suit = shamrock, U+2663 ISOpub" },
1856 { 9829, "hearts","black heart suit = valentine, U+2665 ISOpub" },
1857 { 9830, "diams","black diamond suit, U+2666 ISOpub" },
1861 /************************************************************************
1863 * Commodity functions to handle entities *
1865 ************************************************************************/
1868 * Macro used to grow the current buffer.
1870 #define growBuffer(buffer) { \
1872 buffer##_size *= 2; \
1873 tmp = (xmlChar *) xmlRealloc(buffer, buffer##_size * sizeof(xmlChar)); \
1874 if (tmp == NULL) { \
1875 htmlErrMemory(ctxt, "growing buffer\n"); \
1884 * @name: the entity name
1886 * Lookup the given entity in EntitiesTable
1888 * TODO: the linear scan is really ugly, an hash table is really needed.
1890 * Returns the associated htmlEntityDescPtr if found, NULL otherwise.
1892 const htmlEntityDesc *
1893 htmlEntityLookup(const xmlChar *name) {
1896 for (i = 0;i < (sizeof(html40EntitiesTable)/
1897 sizeof(html40EntitiesTable[0]));i++) {
1898 if (xmlStrEqual(name, BAD_CAST html40EntitiesTable[i].name)) {
1899 return((htmlEntityDescPtr) &html40EntitiesTable[i]);
1906 * htmlEntityValueLookup:
1907 * @value: the entity's unicode value
1909 * Lookup the given entity in EntitiesTable
1911 * TODO: the linear scan is really ugly, an hash table is really needed.
1913 * Returns the associated htmlEntityDescPtr if found, NULL otherwise.
1915 const htmlEntityDesc *
1916 htmlEntityValueLookup(unsigned int value) {
1919 for (i = 0;i < (sizeof(html40EntitiesTable)/
1920 sizeof(html40EntitiesTable[0]));i++) {
1921 if (html40EntitiesTable[i].value >= value) {
1922 if (html40EntitiesTable[i].value > value)
1924 return((htmlEntityDescPtr) &html40EntitiesTable[i]);
1932 * @out: a pointer to an array of bytes to store the result
1933 * @outlen: the length of @out
1934 * @in: a pointer to an array of UTF-8 chars
1935 * @inlen: the length of @in
1937 * Take a block of UTF-8 chars in and try to convert it to an ASCII
1938 * plus HTML entities block of chars out.
1940 * Returns 0 if success, -2 if the transcoding fails, or -1 otherwise
1941 * The value of @inlen after return is the number of octets consumed
1942 * as the return value is positive, else unpredictable.
1943 * The value of @outlen after return is the number of octets consumed.
1946 UTF8ToHtml(unsigned char* out, int *outlen,
1947 const unsigned char* in, int *inlen) {
1948 const unsigned char* processed = in;
1949 const unsigned char* outend;
1950 const unsigned char* outstart = out;
1951 const unsigned char* instart = in;
1952 const unsigned char* inend;
1956 if ((out == NULL) || (outlen == NULL) || (inlen == NULL)) return(-1);
1959 * initialization nothing to do
1965 inend = in + (*inlen);
1966 outend = out + (*outlen);
1967 while (in < inend) {
1969 if (d < 0x80) { c= d; trailing= 0; }
1970 else if (d < 0xC0) {
1971 /* trailing byte in leading position */
1972 *outlen = out - outstart;
1973 *inlen = processed - instart;
1975 } else if (d < 0xE0) { c= d & 0x1F; trailing= 1; }
1976 else if (d < 0xF0) { c= d & 0x0F; trailing= 2; }
1977 else if (d < 0xF8) { c= d & 0x07; trailing= 3; }
1979 /* no chance for this in Ascii */
1980 *outlen = out - outstart;
1981 *inlen = processed - instart;
1985 if (inend - in < trailing) {
1989 for ( ; trailing; trailing--) {
1990 if ((in >= inend) || (((d= *in++) & 0xC0) != 0x80))
1996 /* assertion: c is a single UTF-4 value */
1998 if (out + 1 >= outend)
2003 const htmlEntityDesc * ent;
2008 * Try to lookup a predefined HTML entity for it
2011 ent = htmlEntityValueLookup(c);
2013 snprintf(nbuf, sizeof(nbuf), "#%u", c);
2019 if (out + 2 + len >= outend)
2022 memcpy(out, cp, len);
2028 *outlen = out - outstart;
2029 *inlen = processed - instart;
2034 * htmlEncodeEntities:
2035 * @out: a pointer to an array of bytes to store the result
2036 * @outlen: the length of @out
2037 * @in: a pointer to an array of UTF-8 chars
2038 * @inlen: the length of @in
2039 * @quoteChar: the quote character to escape (' or ") or zero.
2041 * Take a block of UTF-8 chars in and try to convert it to an ASCII
2042 * plus HTML entities block of chars out.
2044 * Returns 0 if success, -2 if the transcoding fails, or -1 otherwise
2045 * The value of @inlen after return is the number of octets consumed
2046 * as the return value is positive, else unpredictable.
2047 * The value of @outlen after return is the number of octets consumed.
2050 htmlEncodeEntities(unsigned char* out, int *outlen,
2051 const unsigned char* in, int *inlen, int quoteChar) {
2052 const unsigned char* processed = in;
2053 const unsigned char* outend;
2054 const unsigned char* outstart = out;
2055 const unsigned char* instart = in;
2056 const unsigned char* inend;
2060 if ((out == NULL) || (outlen == NULL) || (inlen == NULL) || (in == NULL))
2062 outend = out + (*outlen);
2063 inend = in + (*inlen);
2064 while (in < inend) {
2066 if (d < 0x80) { c= d; trailing= 0; }
2067 else if (d < 0xC0) {
2068 /* trailing byte in leading position */
2069 *outlen = out - outstart;
2070 *inlen = processed - instart;
2072 } else if (d < 0xE0) { c= d & 0x1F; trailing= 1; }
2073 else if (d < 0xF0) { c= d & 0x0F; trailing= 2; }
2074 else if (d < 0xF8) { c= d & 0x07; trailing= 3; }
2076 /* no chance for this in Ascii */
2077 *outlen = out - outstart;
2078 *inlen = processed - instart;
2082 if (inend - in < trailing)
2085 while (trailing--) {
2086 if (((d= *in++) & 0xC0) != 0x80) {
2087 *outlen = out - outstart;
2088 *inlen = processed - instart;
2095 /* assertion: c is a single UTF-4 value */
2096 if ((c < 0x80) && (c != (unsigned int) quoteChar) &&
2097 (c != '&') && (c != '<') && (c != '>')) {
2102 const htmlEntityDesc * ent;
2108 * Try to lookup a predefined HTML entity for it
2110 ent = htmlEntityValueLookup(c);
2112 snprintf(nbuf, sizeof(nbuf), "#%u", c);
2118 if (out + 2 + len > outend)
2121 memcpy(out, cp, len);
2127 *outlen = out - outstart;
2128 *inlen = processed - instart;
2132 /************************************************************************
2134 * Commodity functions to handle streams *
2136 ************************************************************************/
2139 * htmlNewInputStream:
2140 * @ctxt: an HTML parser context
2142 * Create a new input stream structure
2143 * Returns the new input stream or NULL
2145 static htmlParserInputPtr
2146 htmlNewInputStream(htmlParserCtxtPtr ctxt) {
2147 htmlParserInputPtr input;
2149 input = (xmlParserInputPtr) xmlMalloc(sizeof(htmlParserInput));
2150 if (input == NULL) {
2151 htmlErrMemory(ctxt, "couldn't allocate a new input stream\n");
2154 memset(input, 0, sizeof(htmlParserInput));
2155 input->filename = NULL;
2156 input->directory = NULL;
2164 input->version = NULL;
2165 input->consumed = 0;
2171 /************************************************************************
2173 * Commodity functions, cleanup needed ? *
2175 ************************************************************************/
2177 * all tags allowing pc data from the html 4.01 loose dtd
2178 * NOTE: it might be more apropriate to integrate this information
2179 * into the html40ElementTable array but I don't want to risk any
2180 * binary incomptibility
2182 static const char *allowPCData[] = {
2183 "a", "abbr", "acronym", "address", "applet", "b", "bdo", "big",
2184 "blockquote", "body", "button", "caption", "center", "cite", "code",
2185 "dd", "del", "dfn", "div", "dt", "em", "font", "form", "h1", "h2",
2186 "h3", "h4", "h5", "h6", "i", "iframe", "ins", "kbd", "label", "legend",
2187 "li", "noframes", "noscript", "object", "p", "pre", "q", "s", "samp",
2188 "small", "span", "strike", "strong", "td", "th", "tt", "u", "var"
2193 * @ctxt: an HTML parser context
2195 * @len: the size of @str
2197 * Is this a sequence of blank chars that one can ignore ?
2199 * Returns 1 if ignorable 0 otherwise.
2202 static int areBlanks(htmlParserCtxtPtr ctxt, const xmlChar *str, int len) {
2205 xmlNodePtr lastChild;
2208 for (j = 0;j < len;j++)
2209 if (!(IS_BLANK_CH(str[j]))) return(0);
2211 if (CUR == 0) return(1);
2212 if (CUR != '<') return(0);
2213 if (ctxt->name == NULL)
2215 if (xmlStrEqual(ctxt->name, BAD_CAST"html"))
2217 if (xmlStrEqual(ctxt->name, BAD_CAST"head"))
2220 /* Only strip CDATA children of the body tag for strict HTML DTDs */
2221 if (xmlStrEqual(ctxt->name, BAD_CAST "body") && ctxt->myDoc != NULL) {
2222 dtd = xmlGetIntSubset(ctxt->myDoc);
2223 if (dtd != NULL && dtd->ExternalID != NULL) {
2224 if (!xmlStrcasecmp(dtd->ExternalID, BAD_CAST "-//W3C//DTD HTML 4.01//EN") ||
2225 !xmlStrcasecmp(dtd->ExternalID, BAD_CAST "-//W3C//DTD HTML 4//EN"))
2230 if (ctxt->node == NULL) return(0);
2231 lastChild = xmlGetLastChild(ctxt->node);
2232 while ((lastChild) && (lastChild->type == XML_COMMENT_NODE))
2233 lastChild = lastChild->prev;
2234 if (lastChild == NULL) {
2235 if ((ctxt->node->type != XML_ELEMENT_NODE) &&
2236 (ctxt->node->content != NULL)) return(0);
2237 /* keep ws in constructs like ...<b> </b>...
2238 for all tags "b" allowing PCDATA */
2239 for ( i = 0; i < sizeof(allowPCData)/sizeof(allowPCData[0]); i++ ) {
2240 if ( xmlStrEqual(ctxt->name, BAD_CAST allowPCData[i]) ) {
2244 } else if (xmlNodeIsText(lastChild)) {
2247 /* keep ws in constructs like <p><b>xy</b> <i>z</i><p>
2248 for all tags "p" allowing PCDATA */
2249 for ( i = 0; i < sizeof(allowPCData)/sizeof(allowPCData[0]); i++ ) {
2250 if ( xmlStrEqual(lastChild->name, BAD_CAST allowPCData[i]) ) {
2260 * @URI: URI for the dtd, or NULL
2261 * @ExternalID: the external ID of the DTD, or NULL
2263 * Creates a new HTML document without a DTD node if @URI and @ExternalID
2266 * Returns a new document, do not initialize the DTD if not provided
2269 htmlNewDocNoDtD(const xmlChar *URI, const xmlChar *ExternalID) {
2273 * Allocate a new document and fill the fields.
2275 cur = (xmlDocPtr) xmlMalloc(sizeof(xmlDoc));
2277 htmlErrMemory(NULL, "HTML document creation failed\n");
2280 memset(cur, 0, sizeof(xmlDoc));
2282 cur->type = XML_HTML_DOCUMENT_NODE;
2283 cur->version = NULL;
2284 cur->intSubset = NULL;
2287 cur->children = NULL;
2288 cur->extSubset = NULL;
2290 cur->encoding = NULL;
2291 cur->standalone = 1;
2292 cur->compression = 0;
2295 cur->_private = NULL;
2296 cur->charset = XML_CHAR_ENCODING_UTF8;
2297 cur->properties = XML_DOC_HTML | XML_DOC_USERBUILT;
2298 if ((ExternalID != NULL) ||
2300 xmlCreateIntSubset(cur, BAD_CAST "html", ExternalID, URI);
2306 * @URI: URI for the dtd, or NULL
2307 * @ExternalID: the external ID of the DTD, or NULL
2309 * Creates a new HTML document
2311 * Returns a new document
2314 htmlNewDoc(const xmlChar *URI, const xmlChar *ExternalID) {
2315 if ((URI == NULL) && (ExternalID == NULL))
2316 return(htmlNewDocNoDtD(
2317 BAD_CAST "http://www.w3.org/TR/REC-html40/loose.dtd",
2318 BAD_CAST "-//W3C//DTD HTML 4.0 Transitional//EN"));
2320 return(htmlNewDocNoDtD(URI, ExternalID));
2324 /************************************************************************
2326 * The parser itself *
2327 * Relates to http://www.w3.org/TR/html40 *
2329 ************************************************************************/
2331 /************************************************************************
2333 * The parser itself *
2335 ************************************************************************/
2337 static const xmlChar * htmlParseNameComplex(xmlParserCtxtPtr ctxt);
2340 * htmlParseHTMLName:
2341 * @ctxt: an HTML parser context
2343 * parse an HTML tag or attribute name, note that we convert it to lowercase
2344 * since HTML names are not case-sensitive.
2346 * Returns the Tag Name parsed or NULL
2349 static const xmlChar *
2350 htmlParseHTMLName(htmlParserCtxtPtr ctxt) {
2352 xmlChar loc[HTML_PARSER_BUFFER_SIZE];
2354 if (!IS_ASCII_LETTER(CUR) && (CUR != '_') &&
2355 (CUR != ':') && (CUR != '.')) return(NULL);
2357 while ((i < HTML_PARSER_BUFFER_SIZE) &&
2358 ((IS_ASCII_LETTER(CUR)) || (IS_ASCII_DIGIT(CUR)) ||
2359 (CUR == ':') || (CUR == '-') || (CUR == '_') ||
2361 if ((CUR >= 'A') && (CUR <= 'Z')) loc[i] = CUR + 0x20;
2368 return(xmlDictLookup(ctxt->dict, loc, i));
2373 * htmlParseHTMLName_nonInvasive:
2374 * @ctxt: an HTML parser context
2376 * parse an HTML tag or attribute name, note that we convert it to lowercase
2377 * since HTML names are not case-sensitive, this doesn't consume the data
2378 * from the stream, it's a look-ahead
2380 * Returns the Tag Name parsed or NULL
2383 static const xmlChar *
2384 htmlParseHTMLName_nonInvasive(htmlParserCtxtPtr ctxt) {
2386 xmlChar loc[HTML_PARSER_BUFFER_SIZE];
2388 if (!IS_ASCII_LETTER(NXT(1)) && (NXT(1) != '_') &&
2389 (NXT(1) != ':')) return(NULL);
2391 while ((i < HTML_PARSER_BUFFER_SIZE) &&
2392 ((IS_ASCII_LETTER(NXT(1+i))) || (IS_ASCII_DIGIT(NXT(1+i))) ||
2393 (NXT(1+i) == ':') || (NXT(1+i) == '-') || (NXT(1+i) == '_'))) {
2394 if ((NXT(1+i) >= 'A') && (NXT(1+i) <= 'Z')) loc[i] = NXT(1+i) + 0x20;
2395 else loc[i] = NXT(1+i);
2399 return(xmlDictLookup(ctxt->dict, loc, i));
2405 * @ctxt: an HTML parser context
2407 * parse an HTML name, this routine is case sensitive.
2409 * Returns the Name parsed or NULL
2412 static const xmlChar *
2413 htmlParseName(htmlParserCtxtPtr ctxt) {
2421 * Accelerator for simple ASCII names
2423 in = ctxt->input->cur;
2424 if (((*in >= 0x61) && (*in <= 0x7A)) ||
2425 ((*in >= 0x41) && (*in <= 0x5A)) ||
2426 (*in == '_') || (*in == ':')) {
2428 while (((*in >= 0x61) && (*in <= 0x7A)) ||
2429 ((*in >= 0x41) && (*in <= 0x5A)) ||
2430 ((*in >= 0x30) && (*in <= 0x39)) ||
2431 (*in == '_') || (*in == '-') ||
2432 (*in == ':') || (*in == '.'))
2434 if ((*in > 0) && (*in < 0x80)) {
2435 count = in - ctxt->input->cur;
2436 ret = xmlDictLookup(ctxt->dict, ctxt->input->cur, count);
2437 ctxt->input->cur = in;
2438 ctxt->nbChars += count;
2439 ctxt->input->col += count;
2443 return(htmlParseNameComplex(ctxt));
2446 static const xmlChar *
2447 htmlParseNameComplex(xmlParserCtxtPtr ctxt) {
2453 * Handler for more complex cases
2457 if ((c == ' ') || (c == '>') || (c == '/') || /* accelerators */
2458 (!IS_LETTER(c) && (c != '_') &&
2463 while ((c != ' ') && (c != '>') && (c != '/') && /* test bigname.xml */
2464 ((IS_LETTER(c)) || (IS_DIGIT(c)) ||
2465 (c == '.') || (c == '-') ||
2466 (c == '_') || (c == ':') ||
2467 (IS_COMBINING(c)) ||
2468 (IS_EXTENDER(c)))) {
2469 if (count++ > 100) {
2477 return(xmlDictLookup(ctxt->dict, ctxt->input->cur - len, len));
2482 * htmlParseHTMLAttribute:
2483 * @ctxt: an HTML parser context
2484 * @stop: a char stop value
2486 * parse an HTML attribute value till the stop (quote), if
2487 * stop is 0 then it stops at the first space
2489 * Returns the attribute parsed or NULL
2493 htmlParseHTMLAttribute(htmlParserCtxtPtr ctxt, const xmlChar stop) {
2494 xmlChar *buffer = NULL;
2495 int buffer_size = 0;
2496 xmlChar *out = NULL;
2497 const xmlChar *name = NULL;
2498 const xmlChar *cur = NULL;
2499 const htmlEntityDesc * ent;
2502 * allocate a translation buffer.
2504 buffer_size = HTML_PARSER_BUFFER_SIZE;
2505 buffer = (xmlChar *) xmlMallocAtomic(buffer_size * sizeof(xmlChar));
2506 if (buffer == NULL) {
2507 htmlErrMemory(ctxt, "buffer allocation failed\n");
2513 * Ok loop until we reach one of the ending chars
2515 while ((CUR != 0) && (CUR != stop)) {
2516 if ((stop == 0) && (CUR == '>')) break;
2517 if ((stop == 0) && (IS_BLANK_CH(CUR))) break;
2519 if (NXT(1) == '#') {
2523 c = htmlParseCharRef(ctxt);
2525 { *out++ = c; bits= -6; }
2527 { *out++ =((c >> 6) & 0x1F) | 0xC0; bits= 0; }
2528 else if (c < 0x10000)
2529 { *out++ =((c >> 12) & 0x0F) | 0xE0; bits= 6; }
2531 { *out++ =((c >> 18) & 0x07) | 0xF0; bits= 12; }
2533 for ( ; bits >= 0; bits-= 6) {
2534 *out++ = ((c >> bits) & 0x3F) | 0x80;
2537 if (out - buffer > buffer_size - 100) {
2538 int indx = out - buffer;
2541 out = &buffer[indx];
2544 ent = htmlParseEntityRef(ctxt, &name);
2547 if (out - buffer > buffer_size - 100) {
2548 int indx = out - buffer;
2551 out = &buffer[indx];
2553 } else if (ent == NULL) {
2557 if (out - buffer > buffer_size - 100) {
2558 int indx = out - buffer;
2561 out = &buffer[indx];
2569 if (out - buffer > buffer_size - 100) {
2570 int indx = out - buffer;
2573 out = &buffer[indx];
2577 { *out++ = c; bits= -6; }
2579 { *out++ =((c >> 6) & 0x1F) | 0xC0; bits= 0; }
2580 else if (c < 0x10000)
2581 { *out++ =((c >> 12) & 0x0F) | 0xE0; bits= 6; }
2583 { *out++ =((c >> 18) & 0x07) | 0xF0; bits= 12; }
2585 for ( ; bits >= 0; bits-= 6) {
2586 *out++ = ((c >> bits) & 0x3F) | 0x80;
2594 if (out - buffer > buffer_size - 100) {
2595 int indx = out - buffer;
2598 out = &buffer[indx];
2602 { *out++ = c; bits= -6; }
2604 { *out++ =((c >> 6) & 0x1F) | 0xC0; bits= 0; }
2605 else if (c < 0x10000)
2606 { *out++ =((c >> 12) & 0x0F) | 0xE0; bits= 6; }
2608 { *out++ =((c >> 18) & 0x07) | 0xF0; bits= 12; }
2610 for ( ; bits >= 0; bits-= 6) {
2611 *out++ = ((c >> bits) & 0x3F) | 0x80;
2621 * htmlParseEntityRef:
2622 * @ctxt: an HTML parser context
2623 * @str: location to store the entity name
2625 * parse an HTML ENTITY references
2627 * [68] EntityRef ::= '&' Name ';'
2629 * Returns the associated htmlEntityDescPtr if found, or NULL otherwise,
2630 * if non-NULL *str will have to be freed by the caller.
2632 const htmlEntityDesc *
2633 htmlParseEntityRef(htmlParserCtxtPtr ctxt, const xmlChar **str) {
2634 const xmlChar *name;
2635 const htmlEntityDesc * ent = NULL;
2637 if (str != NULL) *str = NULL;
2638 if ((ctxt == NULL) || (ctxt->input == NULL)) return(NULL);
2642 name = htmlParseName(ctxt);
2644 htmlParseErr(ctxt, XML_ERR_NAME_REQUIRED,
2645 "htmlParseEntityRef: no name\n", NULL, NULL);
2653 * Lookup the entity in the table.
2655 ent = htmlEntityLookup(name);
2656 if (ent != NULL) /* OK that's ugly !!! */
2659 htmlParseErr(ctxt, XML_ERR_ENTITYREF_SEMICOL_MISSING,
2660 "htmlParseEntityRef: expecting ';'\n",
2671 * htmlParseAttValue:
2672 * @ctxt: an HTML parser context
2674 * parse a value for an attribute
2675 * Note: the parser won't do substitution of entities here, this
2676 * will be handled later in xmlStringGetNodeList, unless it was
2677 * asked for ctxt->replaceEntities != 0
2679 * Returns the AttValue parsed or NULL.
2683 htmlParseAttValue(htmlParserCtxtPtr ctxt) {
2684 xmlChar *ret = NULL;
2688 ret = htmlParseHTMLAttribute(ctxt, '"');
2690 htmlParseErr(ctxt, XML_ERR_ATTRIBUTE_NOT_FINISHED,
2691 "AttValue: \" expected\n", NULL, NULL);
2694 } else if (CUR == '\'') {
2696 ret = htmlParseHTMLAttribute(ctxt, '\'');
2698 htmlParseErr(ctxt, XML_ERR_ATTRIBUTE_NOT_FINISHED,
2699 "AttValue: ' expected\n", NULL, NULL);
2704 * That's an HTMLism, the attribute value may not be quoted
2706 ret = htmlParseHTMLAttribute(ctxt, 0);
2708 htmlParseErr(ctxt, XML_ERR_ATTRIBUTE_WITHOUT_VALUE,
2709 "AttValue: no value found\n", NULL, NULL);
2716 * htmlParseSystemLiteral:
2717 * @ctxt: an HTML parser context
2719 * parse an HTML Literal
2721 * [11] SystemLiteral ::= ('"' [^"]* '"') | ("'" [^']* "'")
2723 * Returns the SystemLiteral parsed or NULL
2727 htmlParseSystemLiteral(htmlParserCtxtPtr ctxt) {
2729 xmlChar *ret = NULL;
2734 while ((IS_CHAR_CH(CUR)) && (CUR != '"'))
2736 if (!IS_CHAR_CH(CUR)) {
2737 htmlParseErr(ctxt, XML_ERR_LITERAL_NOT_FINISHED,
2738 "Unfinished SystemLiteral\n", NULL, NULL);
2740 ret = xmlStrndup(q, CUR_PTR - q);
2743 } else if (CUR == '\'') {
2746 while ((IS_CHAR_CH(CUR)) && (CUR != '\''))
2748 if (!IS_CHAR_CH(CUR)) {
2749 htmlParseErr(ctxt, XML_ERR_LITERAL_NOT_FINISHED,
2750 "Unfinished SystemLiteral\n", NULL, NULL);
2752 ret = xmlStrndup(q, CUR_PTR - q);
2756 htmlParseErr(ctxt, XML_ERR_LITERAL_NOT_STARTED,
2757 " or ' expected\n", NULL, NULL);
2764 * htmlParsePubidLiteral:
2765 * @ctxt: an HTML parser context
2767 * parse an HTML public literal
2769 * [12] PubidLiteral ::= '"' PubidChar* '"' | "'" (PubidChar - "'")* "'"
2771 * Returns the PubidLiteral parsed or NULL.
2775 htmlParsePubidLiteral(htmlParserCtxtPtr ctxt) {
2777 xmlChar *ret = NULL;
2779 * Name ::= (Letter | '_') (NameChar)*
2784 while (IS_PUBIDCHAR_CH(CUR)) NEXT;
2786 htmlParseErr(ctxt, XML_ERR_LITERAL_NOT_FINISHED,
2787 "Unfinished PubidLiteral\n", NULL, NULL);
2789 ret = xmlStrndup(q, CUR_PTR - q);
2792 } else if (CUR == '\'') {
2795 while ((IS_PUBIDCHAR_CH(CUR)) && (CUR != '\''))
2798 htmlParseErr(ctxt, XML_ERR_LITERAL_NOT_FINISHED,
2799 "Unfinished PubidLiteral\n", NULL, NULL);
2801 ret = xmlStrndup(q, CUR_PTR - q);
2805 htmlParseErr(ctxt, XML_ERR_LITERAL_NOT_STARTED,
2806 "PubidLiteral \" or ' expected\n", NULL, NULL);
2814 * @ctxt: an HTML parser context
2816 * parse the content of an HTML SCRIPT or STYLE element
2817 * http://www.w3.org/TR/html4/sgml/dtd.html#Script
2818 * http://www.w3.org/TR/html4/sgml/dtd.html#StyleSheet
2819 * http://www.w3.org/TR/html4/types.html#type-script
2820 * http://www.w3.org/TR/html4/types.html#h-6.15
2821 * http://www.w3.org/TR/html4/appendix/notes.html#h-B.3.2.1
2823 * Script data ( %Script; in the DTD) can be the content of the SCRIPT
2824 * element and the value of intrinsic event attributes. User agents must
2825 * not evaluate script data as HTML markup but instead must pass it on as
2826 * data to a script engine.
2828 * - The content is passed like CDATA
2829 * - the attributes for style and scripting "onXXX" are also described
2830 * as CDATA but SGML allows entities references in attributes so their
2831 * processing is identical as other attributes
2834 htmlParseScript(htmlParserCtxtPtr ctxt) {
2835 xmlChar buf[HTML_PARSER_BIG_BUFFER_SIZE + 5];
2841 while (IS_CHAR_CH(cur)) {
2842 if ((cur == '<') && (NXT(1) == '/')) {
2844 * One should break here, the specification is clear:
2845 * Authors should therefore escape "</" within the content.
2846 * Escape mechanisms are specific to each scripting or
2847 * style sheet language.
2849 * In recovery mode, only break if end tag match the
2850 * current tag, effectively ignoring all tags inside the
2851 * script/style block and treating the entire block as
2854 if (ctxt->recovery) {
2855 if (xmlStrncasecmp(ctxt->name, ctxt->input->cur+2,
2856 xmlStrlen(ctxt->name)) == 0)
2860 htmlParseErr(ctxt, XML_ERR_TAG_NAME_MISMATCH,
2861 "Element %s embeds close tag\n",
2865 if (((NXT(2) >= 'A') && (NXT(2) <= 'Z')) ||
2866 ((NXT(2) >= 'a') && (NXT(2) <= 'z')))
2872 COPY_BUF(l,buf,nbchar,cur);
2873 if (nbchar >= HTML_PARSER_BIG_BUFFER_SIZE) {
2874 if (ctxt->sax->cdataBlock!= NULL) {
2876 * Insert as CDATA, which is the same as HTML_PRESERVE_NODE
2878 ctxt->sax->cdataBlock(ctxt->userData, buf, nbchar);
2879 } else if (ctxt->sax->characters != NULL) {
2880 ctxt->sax->characters(ctxt->userData, buf, nbchar);
2889 if ((!(IS_CHAR_CH(cur))) && (!((cur == 0) && (ctxt->progressive)))) {
2890 htmlParseErrInt(ctxt, XML_ERR_INVALID_CHAR,
2891 "Invalid char in CDATA 0x%X\n", cur);
2892 if (ctxt->input->cur < ctxt->input->end) {
2897 if ((nbchar != 0) && (ctxt->sax != NULL) && (!ctxt->disableSAX)) {
2898 if (ctxt->sax->cdataBlock!= NULL) {
2900 * Insert as CDATA, which is the same as HTML_PRESERVE_NODE
2902 ctxt->sax->cdataBlock(ctxt->userData, buf, nbchar);
2903 } else if (ctxt->sax->characters != NULL) {
2904 ctxt->sax->characters(ctxt->userData, buf, nbchar);
2911 * htmlParseCharData:
2912 * @ctxt: an HTML parser context
2914 * parse a CharData section.
2915 * if we are within a CDATA section ']]>' marks an end of section.
2917 * [14] CharData ::= [^<&]* - ([^<&]* ']]>' [^<&]*)
2921 htmlParseCharData(htmlParserCtxtPtr ctxt) {
2922 xmlChar buf[HTML_PARSER_BIG_BUFFER_SIZE + 5];
2929 while (((cur != '<') || (ctxt->token == '<')) &&
2930 ((cur != '&') || (ctxt->token == '&')) &&
2932 if (!(IS_CHAR(cur))) {
2933 htmlParseErrInt(ctxt, XML_ERR_INVALID_CHAR,
2934 "Invalid char in CDATA 0x%X\n", cur);
2936 COPY_BUF(l,buf,nbchar,cur);
2938 if (nbchar >= HTML_PARSER_BIG_BUFFER_SIZE) {
2940 * Ok the segment is to be consumed as chars.
2942 if ((ctxt->sax != NULL) && (!ctxt->disableSAX)) {
2943 if (areBlanks(ctxt, buf, nbchar)) {
2944 if (ctxt->sax->ignorableWhitespace != NULL)
2945 ctxt->sax->ignorableWhitespace(ctxt->userData,
2948 htmlCheckParagraph(ctxt);
2949 if (ctxt->sax->characters != NULL)
2950 ctxt->sax->characters(ctxt->userData, buf, nbchar);
2957 if (chunk > HTML_PARSER_BUFFER_SIZE) {
2973 * Ok the segment is to be consumed as chars.
2975 if ((ctxt->sax != NULL) && (!ctxt->disableSAX)) {
2976 if (areBlanks(ctxt, buf, nbchar)) {
2977 if (ctxt->sax->ignorableWhitespace != NULL)
2978 ctxt->sax->ignorableWhitespace(ctxt->userData, buf, nbchar);
2980 htmlCheckParagraph(ctxt);
2981 if (ctxt->sax->characters != NULL)
2982 ctxt->sax->characters(ctxt->userData, buf, nbchar);
2990 ctxt->instate = XML_PARSER_EOF;
2995 * htmlParseExternalID:
2996 * @ctxt: an HTML parser context
2997 * @publicID: a xmlChar** receiving PubidLiteral
2999 * Parse an External ID or a Public ID
3001 * [75] ExternalID ::= 'SYSTEM' S SystemLiteral
3002 * | 'PUBLIC' S PubidLiteral S SystemLiteral
3004 * [83] PublicID ::= 'PUBLIC' S PubidLiteral
3006 * Returns the function returns SystemLiteral and in the second
3007 * case publicID receives PubidLiteral, is strict is off
3008 * it is possible to return NULL and have publicID set.
3012 htmlParseExternalID(htmlParserCtxtPtr ctxt, xmlChar **publicID) {
3013 xmlChar *URI = NULL;
3015 if ((UPPER == 'S') && (UPP(1) == 'Y') &&
3016 (UPP(2) == 'S') && (UPP(3) == 'T') &&
3017 (UPP(4) == 'E') && (UPP(5) == 'M')) {
3019 if (!IS_BLANK_CH(CUR)) {
3020 htmlParseErr(ctxt, XML_ERR_SPACE_REQUIRED,
3021 "Space required after 'SYSTEM'\n", NULL, NULL);
3024 URI = htmlParseSystemLiteral(ctxt);
3026 htmlParseErr(ctxt, XML_ERR_URI_REQUIRED,
3027 "htmlParseExternalID: SYSTEM, no URI\n", NULL, NULL);
3029 } else if ((UPPER == 'P') && (UPP(1) == 'U') &&
3030 (UPP(2) == 'B') && (UPP(3) == 'L') &&
3031 (UPP(4) == 'I') && (UPP(5) == 'C')) {
3033 if (!IS_BLANK_CH(CUR)) {
3034 htmlParseErr(ctxt, XML_ERR_SPACE_REQUIRED,
3035 "Space required after 'PUBLIC'\n", NULL, NULL);
3038 *publicID = htmlParsePubidLiteral(ctxt);
3039 if (*publicID == NULL) {
3040 htmlParseErr(ctxt, XML_ERR_PUBID_REQUIRED,
3041 "htmlParseExternalID: PUBLIC, no Public Identifier\n",
3045 if ((CUR == '"') || (CUR == '\'')) {
3046 URI = htmlParseSystemLiteral(ctxt);
3054 * @ctxt: an XML parser context
3056 * parse an XML Processing Instruction.
3058 * [16] PI ::= '<?' PITarget (S (Char* - (Char* '?>' Char*)))? '?>'
3061 htmlParsePI(htmlParserCtxtPtr ctxt) {
3062 xmlChar *buf = NULL;
3064 int size = HTML_PARSER_BUFFER_SIZE;
3066 const xmlChar *target;
3067 xmlParserInputState state;
3070 if ((RAW == '<') && (NXT(1) == '?')) {
3071 state = ctxt->instate;
3072 ctxt->instate = XML_PARSER_PI;
3074 * this is a Processing Instruction.
3080 * Parse the target name and check for special support like
3083 target = htmlParseName(ctxt);
3084 if (target != NULL) {
3091 if ((ctxt->sax) && (!ctxt->disableSAX) &&
3092 (ctxt->sax->processingInstruction != NULL))
3093 ctxt->sax->processingInstruction(ctxt->userData,
3095 ctxt->instate = state;
3098 buf = (xmlChar *) xmlMallocAtomic(size * sizeof(xmlChar));
3100 htmlErrMemory(ctxt, NULL);
3101 ctxt->instate = state;
3105 if (!IS_BLANK(cur)) {
3106 htmlParseErr(ctxt, XML_ERR_SPACE_REQUIRED,
3107 "ParsePI: PI %s space expected\n", target, NULL);
3111 while (IS_CHAR(cur) && (cur != '>')) {
3112 if (len + 5 >= size) {
3116 tmp = (xmlChar *) xmlRealloc(buf, size * sizeof(xmlChar));
3118 htmlErrMemory(ctxt, NULL);
3120 ctxt->instate = state;
3130 COPY_BUF(l,buf,len,cur);
3141 htmlParseErr(ctxt, XML_ERR_PI_NOT_FINISHED,
3142 "ParsePI: PI %s never end ...\n", target, NULL);
3149 if ((ctxt->sax) && (!ctxt->disableSAX) &&
3150 (ctxt->sax->processingInstruction != NULL))
3151 ctxt->sax->processingInstruction(ctxt->userData,
3156 htmlParseErr(ctxt, XML_ERR_PI_NOT_STARTED,
3157 "PI is not started correctly", NULL, NULL);
3159 ctxt->instate = state;
3165 * @ctxt: an HTML parser context
3167 * Parse an XML (SGML) comment <!-- .... -->
3169 * [15] Comment ::= '<!--' ((Char - '-') | ('-' (Char - '-')))* '-->'
3172 htmlParseComment(htmlParserCtxtPtr ctxt) {
3173 xmlChar *buf = NULL;
3175 int size = HTML_PARSER_BUFFER_SIZE;
3179 xmlParserInputState state;
3182 * Check that there is a comment right here.
3184 if ((RAW != '<') || (NXT(1) != '!') ||
3185 (NXT(2) != '-') || (NXT(3) != '-')) return;
3187 state = ctxt->instate;
3188 ctxt->instate = XML_PARSER_COMMENT;
3191 buf = (xmlChar *) xmlMallocAtomic(size * sizeof(xmlChar));
3193 htmlErrMemory(ctxt, "buffer allocation failed\n");
3194 ctxt->instate = state;
3203 while (IS_CHAR(cur) &&
3205 (r != '-') || (q != '-'))) {
3206 if (len + 5 >= size) {
3210 tmp = (xmlChar *) xmlRealloc(buf, size * sizeof(xmlChar));
3213 htmlErrMemory(ctxt, "growing buffer failed\n");
3214 ctxt->instate = state;
3219 COPY_BUF(ql,buf,len,q);
3233 if (!IS_CHAR(cur)) {
3234 htmlParseErr(ctxt, XML_ERR_COMMENT_NOT_FINISHED,
3235 "Comment not terminated \n<!--%.50s\n", buf, NULL);
3239 if ((ctxt->sax != NULL) && (ctxt->sax->comment != NULL) &&
3240 (!ctxt->disableSAX))
3241 ctxt->sax->comment(ctxt->userData, buf);
3244 ctxt->instate = state;
3249 * @ctxt: an HTML parser context
3251 * parse Reference declarations
3253 * [66] CharRef ::= '&#' [0-9]+ ';' |
3254 * '&#x' [0-9a-fA-F]+ ';'
3256 * Returns the value parsed (as an int)
3259 htmlParseCharRef(htmlParserCtxtPtr ctxt) {
3262 if ((ctxt == NULL) || (ctxt->input == NULL)) {
3263 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
3264 "htmlParseCharRef: context error\n",
3268 if ((CUR == '&') && (NXT(1) == '#') &&
3269 ((NXT(2) == 'x') || NXT(2) == 'X')) {
3271 while (CUR != ';') {
3272 if ((CUR >= '0') && (CUR <= '9'))
3273 val = val * 16 + (CUR - '0');
3274 else if ((CUR >= 'a') && (CUR <= 'f'))
3275 val = val * 16 + (CUR - 'a') + 10;
3276 else if ((CUR >= 'A') && (CUR <= 'F'))
3277 val = val * 16 + (CUR - 'A') + 10;
3279 htmlParseErr(ctxt, XML_ERR_INVALID_HEX_CHARREF,
3280 "htmlParseCharRef: missing semicolon\n",
3288 } else if ((CUR == '&') && (NXT(1) == '#')) {
3290 while (CUR != ';') {
3291 if ((CUR >= '0') && (CUR <= '9'))
3292 val = val * 10 + (CUR - '0');
3294 htmlParseErr(ctxt, XML_ERR_INVALID_DEC_CHARREF,
3295 "htmlParseCharRef: missing semicolon\n",
3304 htmlParseErr(ctxt, XML_ERR_INVALID_CHARREF,
3305 "htmlParseCharRef: invalid value\n", NULL, NULL);
3308 * Check the value IS_CHAR ...
3313 htmlParseErrInt(ctxt, XML_ERR_INVALID_CHAR,
3314 "htmlParseCharRef: invalid xmlChar value %d\n",
3322 * htmlParseDocTypeDecl:
3323 * @ctxt: an HTML parser context
3325 * parse a DOCTYPE declaration
3327 * [28] doctypedecl ::= '<!DOCTYPE' S Name (S ExternalID)? S?
3328 * ('[' (markupdecl | PEReference | S)* ']' S?)? '>'
3332 htmlParseDocTypeDecl(htmlParserCtxtPtr ctxt) {
3333 const xmlChar *name;
3334 xmlChar *ExternalID = NULL;
3335 xmlChar *URI = NULL;
3338 * We know that '<!DOCTYPE' has been detected.
3345 * Parse the DOCTYPE name.
3347 name = htmlParseName(ctxt);
3349 htmlParseErr(ctxt, XML_ERR_NAME_REQUIRED,
3350 "htmlParseDocTypeDecl : no DOCTYPE name !\n",
3354 * Check that upper(name) == "HTML" !!!!!!!!!!!!!
3360 * Check for SystemID and ExternalID
3362 URI = htmlParseExternalID(ctxt, &ExternalID);
3366 * We should be at the end of the DOCTYPE declaration.
3369 htmlParseErr(ctxt, XML_ERR_DOCTYPE_NOT_FINISHED,
3370 "DOCTYPE improperly terminated\n", NULL, NULL);
3371 /* We shouldn't try to resynchronize ... */
3376 * Create or update the document accordingly to the DOCTYPE
3378 if ((ctxt->sax != NULL) && (ctxt->sax->internalSubset != NULL) &&
3379 (!ctxt->disableSAX))
3380 ctxt->sax->internalSubset(ctxt->userData, name, ExternalID, URI);
3383 * Cleanup, since we don't use all those identifiers
3385 if (URI != NULL) xmlFree(URI);
3386 if (ExternalID != NULL) xmlFree(ExternalID);
3390 * htmlParseAttribute:
3391 * @ctxt: an HTML parser context
3392 * @value: a xmlChar ** used to store the value of the attribute
3394 * parse an attribute
3396 * [41] Attribute ::= Name Eq AttValue
3398 * [25] Eq ::= S? '=' S?
3402 * [NS 11] Attribute ::= QName Eq AttValue
3404 * Also the case QName == xmlns:??? is handled independently as a namespace
3407 * Returns the attribute name, and the value in *value.
3410 static const xmlChar *
3411 htmlParseAttribute(htmlParserCtxtPtr ctxt, xmlChar **value) {
3412 const xmlChar *name;
3413 xmlChar *val = NULL;
3416 name = htmlParseHTMLName(ctxt);
3418 htmlParseErr(ctxt, XML_ERR_NAME_REQUIRED,
3419 "error parsing attribute name\n", NULL, NULL);
3430 val = htmlParseAttValue(ctxt);
3438 * htmlCheckEncoding:
3439 * @ctxt: an HTML parser context
3440 * @attvalue: the attribute value
3442 * Checks an http-equiv attribute from a Meta tag to detect
3444 * If a new encoding is detected the parser is switched to decode
3448 htmlCheckEncoding(htmlParserCtxtPtr ctxt, const xmlChar *attvalue) {
3449 const xmlChar *encoding;
3451 if ((ctxt == NULL) || (attvalue == NULL))
3454 /* do not change encoding */
3455 if (ctxt->input->encoding != NULL)
3458 encoding = xmlStrcasestr(attvalue, BAD_CAST"charset=");
3459 if (encoding != NULL) {
3462 encoding = xmlStrcasestr(attvalue, BAD_CAST"charset =");
3463 if (encoding != NULL)
3466 if (encoding != NULL) {
3467 xmlCharEncoding enc;
3468 xmlCharEncodingHandlerPtr handler;
3470 while ((*encoding == ' ') || (*encoding == '\t')) encoding++;
3472 if (ctxt->input->encoding != NULL)
3473 xmlFree((xmlChar *) ctxt->input->encoding);
3474 ctxt->input->encoding = xmlStrdup(encoding);
3476 enc = xmlParseCharEncoding((const char *) encoding);
3478 * registered set of known encodings
3480 if (enc != XML_CHAR_ENCODING_ERROR) {
3481 if (((enc == XML_CHAR_ENCODING_UTF16LE) ||
3482 (enc == XML_CHAR_ENCODING_UTF16BE) ||
3483 (enc == XML_CHAR_ENCODING_UCS4LE) ||
3484 (enc == XML_CHAR_ENCODING_UCS4BE)) &&
3485 (ctxt->input->buf != NULL) &&
3486 (ctxt->input->buf->encoder == NULL)) {
3487 htmlParseErr(ctxt, XML_ERR_INVALID_ENCODING,
3488 "htmlCheckEncoding: wrong encoding meta\n",
3491 xmlSwitchEncoding(ctxt, enc);
3493 ctxt->charset = XML_CHAR_ENCODING_UTF8;
3496 * fallback for unknown encodings
3498 handler = xmlFindCharEncodingHandler((const char *) encoding);
3499 if (handler != NULL) {
3500 xmlSwitchToEncoding(ctxt, handler);
3501 ctxt->charset = XML_CHAR_ENCODING_UTF8;
3503 ctxt->errNo = XML_ERR_UNSUPPORTED_ENCODING;
3507 if ((ctxt->input->buf != NULL) &&
3508 (ctxt->input->buf->encoder != NULL) &&
3509 (ctxt->input->buf->raw != NULL) &&
3510 (ctxt->input->buf->buffer != NULL)) {
3515 * convert as much as possible to the parser reading buffer.
3517 processed = ctxt->input->cur - ctxt->input->base;
3518 xmlBufferShrink(ctxt->input->buf->buffer, processed);
3519 nbchars = xmlCharEncInFunc(ctxt->input->buf->encoder,
3520 ctxt->input->buf->buffer,
3521 ctxt->input->buf->raw);
3523 htmlParseErr(ctxt, XML_ERR_INVALID_ENCODING,
3524 "htmlCheckEncoding: encoder error\n",
3528 ctxt->input->cur = ctxt->input->buf->buffer->content;
3530 &ctxt->input->base[ctxt->input->buf->buffer->use];
3537 * @ctxt: an HTML parser context
3538 * @atts: the attributes values
3540 * Checks an attributes from a Meta tag
3543 htmlCheckMeta(htmlParserCtxtPtr ctxt, const xmlChar **atts) {
3545 const xmlChar *att, *value;
3547 const xmlChar *content = NULL;
3549 if ((ctxt == NULL) || (atts == NULL))
3554 while (att != NULL) {
3556 if ((value != NULL) && (!xmlStrcasecmp(att, BAD_CAST"http-equiv"))
3557 && (!xmlStrcasecmp(value, BAD_CAST"Content-Type")))
3559 else if ((value != NULL) && (!xmlStrcasecmp(att, BAD_CAST"content")))
3563 if ((http) && (content != NULL))
3564 htmlCheckEncoding(ctxt, content);
3569 * htmlParseStartTag:
3570 * @ctxt: an HTML parser context
3572 * parse a start of tag either for rule element or
3573 * EmptyElement. In both case we don't parse the tag closing chars.
3575 * [40] STag ::= '<' Name (S Attribute)* S? '>'
3577 * [44] EmptyElemTag ::= '<' Name (S Attribute)* S? '/>'
3581 * [NS 8] STag ::= '<' QName (S Attribute)* S? '>'
3583 * [NS 10] EmptyElement ::= '<' QName (S Attribute)* S? '/>'
3585 * Returns 0 in case of success, -1 in case of error and 1 if discarded
3589 htmlParseStartTag(htmlParserCtxtPtr ctxt) {
3590 const xmlChar *name;
3591 const xmlChar *attname;
3593 const xmlChar **atts;
3600 if (ctxt->instate == XML_PARSER_EOF)
3602 if ((ctxt == NULL) || (ctxt->input == NULL)) {
3603 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
3604 "htmlParseStartTag: context error\n", NULL, NULL);
3607 if (CUR != '<') return -1;
3611 maxatts = ctxt->maxatts;
3614 name = htmlParseHTMLName(ctxt);
3616 htmlParseErr(ctxt, XML_ERR_NAME_REQUIRED,
3617 "htmlParseStartTag: invalid element name\n",
3619 /* Dump the bogus tag like browsers do */
3620 while ((IS_CHAR_CH(CUR)) && (CUR != '>') &&
3621 (ctxt->instate != XML_PARSER_EOF))
3625 if (xmlStrEqual(name, BAD_CAST"meta"))
3629 * Check for auto-closure of HTML elements.
3631 htmlAutoClose(ctxt, name);
3634 * Check for implied HTML elements.
3636 htmlCheckImplied(ctxt, name);
3639 * Avoid html at any level > 0, head at any level != 1
3640 * or any attempt to recurse body
3642 if ((ctxt->nameNr > 0) && (xmlStrEqual(name, BAD_CAST"html"))) {
3643 htmlParseErr(ctxt, XML_HTML_STRUCURE_ERROR,
3644 "htmlParseStartTag: misplaced <html> tag\n",
3649 if ((ctxt->nameNr != 1) &&
3650 (xmlStrEqual(name, BAD_CAST"head"))) {
3651 htmlParseErr(ctxt, XML_HTML_STRUCURE_ERROR,
3652 "htmlParseStartTag: misplaced <head> tag\n",
3657 if (xmlStrEqual(name, BAD_CAST"body")) {
3659 for (indx = 0;indx < ctxt->nameNr;indx++) {
3660 if (xmlStrEqual(ctxt->nameTab[indx], BAD_CAST"body")) {
3661 htmlParseErr(ctxt, XML_HTML_STRUCURE_ERROR,
3662 "htmlParseStartTag: misplaced <body> tag\n",
3671 * Now parse the attributes, it ends up with the ending
3676 while ((IS_CHAR_CH(CUR)) &&
3678 ((CUR != '/') || (NXT(1) != '>'))) {
3679 long cons = ctxt->nbChars;
3682 attname = htmlParseAttribute(ctxt, &attvalue);
3683 if (attname != NULL) {
3686 * Well formedness requires at most one declaration of an attribute
3688 for (i = 0; i < nbatts;i += 2) {
3689 if (xmlStrEqual(atts[i], attname)) {
3690 htmlParseErr(ctxt, XML_ERR_ATTRIBUTE_REDEFINED,
3691 "Attribute %s redefined\n", attname, NULL);
3692 if (attvalue != NULL)
3699 * Add the pair to atts
3702 maxatts = 22; /* allow for 10 attrs by default */
3703 atts = (const xmlChar **)
3704 xmlMalloc(maxatts * sizeof(xmlChar *));
3706 htmlErrMemory(ctxt, NULL);
3707 if (attvalue != NULL)
3712 ctxt->maxatts = maxatts;
3713 } else if (nbatts + 4 > maxatts) {
3717 n = (const xmlChar **) xmlRealloc((void *) atts,
3718 maxatts * sizeof(const xmlChar *));
3720 htmlErrMemory(ctxt, NULL);
3721 if (attvalue != NULL)
3727 ctxt->maxatts = maxatts;
3729 atts[nbatts++] = attname;
3730 atts[nbatts++] = attvalue;
3731 atts[nbatts] = NULL;
3732 atts[nbatts + 1] = NULL;
3735 if (attvalue != NULL)
3737 /* Dump the bogus attribute string up to the next blank or
3738 * the end of the tag. */
3739 while ((IS_CHAR_CH(CUR)) &&
3740 !(IS_BLANK_CH(CUR)) && (CUR != '>') &&
3741 ((CUR != '/') || (NXT(1) != '>')))
3747 if (cons == ctxt->nbChars) {
3748 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
3749 "htmlParseStartTag: problem parsing attributes\n",
3756 * Handle specific association to the META tag
3758 if (meta && (nbatts != 0))
3759 htmlCheckMeta(ctxt, atts);
3762 * SAX: Start of Element !
3765 htmlnamePush(ctxt, name);
3766 if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL)) {
3768 ctxt->sax->startElement(ctxt->userData, name, atts);
3770 ctxt->sax->startElement(ctxt->userData, name, NULL);
3775 for (i = 1;i < nbatts;i += 2) {
3776 if (atts[i] != NULL)
3777 xmlFree((xmlChar *) atts[i]);
3786 * @ctxt: an HTML parser context
3788 * parse an end of tag
3790 * [42] ETag ::= '</' Name S? '>'
3794 * [NS 9] ETag ::= '</' QName S? '>'
3796 * Returns 1 if the current level should be closed.
3800 htmlParseEndTag(htmlParserCtxtPtr ctxt)
3802 const xmlChar *name;
3803 const xmlChar *oldname;
3806 if ((CUR != '<') || (NXT(1) != '/')) {
3807 htmlParseErr(ctxt, XML_ERR_LTSLASH_REQUIRED,
3808 "htmlParseEndTag: '</' not found\n", NULL, NULL);
3813 name = htmlParseHTMLName(ctxt);
3817 * We should definitely be at the ending "S? '>'" part
3820 if ((!IS_CHAR_CH(CUR)) || (CUR != '>')) {
3821 htmlParseErr(ctxt, XML_ERR_GT_REQUIRED,
3822 "End tag : expected '>'\n", NULL, NULL);
3823 if (ctxt->recovery) {
3825 * We're not at the ending > !!
3826 * Error, unless in recover mode where we search forwards
3829 while (CUR != '\0' && CUR != '>') NEXT;
3836 * if we ignored misplaced tags in htmlParseStartTag don't pop them
3839 if ((ctxt->depth > 0) &&
3840 (xmlStrEqual(name, BAD_CAST "html") ||
3841 xmlStrEqual(name, BAD_CAST "body") ||
3842 xmlStrEqual(name, BAD_CAST "head"))) {
3848 * If the name read is not one of the element in the parsing stack
3849 * then return, it's just an error.
3851 for (i = (ctxt->nameNr - 1); i >= 0; i--) {
3852 if (xmlStrEqual(name, ctxt->nameTab[i]))
3856 htmlParseErr(ctxt, XML_ERR_TAG_NAME_MISMATCH,
3857 "Unexpected end tag : %s\n", name, NULL);
3863 * Check for auto-closure of HTML elements.
3866 htmlAutoCloseOnClose(ctxt, name);
3869 * Well formedness constraints, opening and closing must match.
3870 * With the exception that the autoclose may have popped stuff out
3873 if (!xmlStrEqual(name, ctxt->name)) {
3874 if ((ctxt->name != NULL) && (!xmlStrEqual(ctxt->name, name))) {
3875 htmlParseErr(ctxt, XML_ERR_TAG_NAME_MISMATCH,
3876 "Opening and ending tag mismatch: %s and %s\n",
3884 oldname = ctxt->name;
3885 if ((oldname != NULL) && (xmlStrEqual(oldname, name))) {
3886 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
3887 ctxt->sax->endElement(ctxt->userData, name);
3899 * htmlParseReference:
3900 * @ctxt: an HTML parser context
3902 * parse and handle entity references in content,
3903 * this will end-up in a call to character() since this is either a
3904 * CharRef, or a predefined entity.
3907 htmlParseReference(htmlParserCtxtPtr ctxt) {
3908 const htmlEntityDesc * ent;
3910 const xmlChar *name;
3911 if (CUR != '&') return;
3913 if (NXT(1) == '#') {
3917 c = htmlParseCharRef(ctxt);
3921 if (c < 0x80) { out[i++]= c; bits= -6; }
3922 else if (c < 0x800) { out[i++]=((c >> 6) & 0x1F) | 0xC0; bits= 0; }
3923 else if (c < 0x10000) { out[i++]=((c >> 12) & 0x0F) | 0xE0; bits= 6; }
3924 else { out[i++]=((c >> 18) & 0x07) | 0xF0; bits= 12; }
3926 for ( ; bits >= 0; bits-= 6) {
3927 out[i++]= ((c >> bits) & 0x3F) | 0x80;
3931 htmlCheckParagraph(ctxt);
3932 if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
3933 ctxt->sax->characters(ctxt->userData, out, i);
3935 ent = htmlParseEntityRef(ctxt, &name);
3937 htmlCheckParagraph(ctxt);
3938 if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
3939 ctxt->sax->characters(ctxt->userData, BAD_CAST "&", 1);
3942 if ((ent == NULL) || !(ent->value > 0)) {
3943 htmlCheckParagraph(ctxt);
3944 if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL)) {
3945 ctxt->sax->characters(ctxt->userData, BAD_CAST "&", 1);
3946 ctxt->sax->characters(ctxt->userData, name, xmlStrlen(name));
3947 /* ctxt->sax->characters(ctxt->userData, BAD_CAST ";", 1); */
3955 { out[i++]= c; bits= -6; }
3957 { out[i++]=((c >> 6) & 0x1F) | 0xC0; bits= 0; }
3958 else if (c < 0x10000)
3959 { out[i++]=((c >> 12) & 0x0F) | 0xE0; bits= 6; }
3961 { out[i++]=((c >> 18) & 0x07) | 0xF0; bits= 12; }
3963 for ( ; bits >= 0; bits-= 6) {
3964 out[i++]= ((c >> bits) & 0x3F) | 0x80;
3968 htmlCheckParagraph(ctxt);
3969 if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
3970 ctxt->sax->characters(ctxt->userData, out, i);
3977 * @ctxt: an HTML parser context
3979 * Parse a content: comment, sub-element, reference or text.
3980 * Kept for compatibility with old code
3984 htmlParseContent(htmlParserCtxtPtr ctxt) {
3985 xmlChar *currentNode;
3987 const xmlChar *name;
3989 currentNode = xmlStrdup(ctxt->name);
3990 depth = ctxt->nameNr;
3992 long cons = ctxt->nbChars;
3996 if (ctxt->instate == XML_PARSER_EOF)
4000 * Our tag or one of it's parent or children is ending.
4002 if ((CUR == '<') && (NXT(1) == '/')) {
4003 if (htmlParseEndTag(ctxt) &&
4004 ((currentNode != NULL) || (ctxt->nameNr == 0))) {
4005 if (currentNode != NULL)
4006 xmlFree(currentNode);
4009 continue; /* while */
4012 else if ((CUR == '<') &&
4013 ((IS_ASCII_LETTER(NXT(1))) ||
4014 (NXT(1) == '_') || (NXT(1) == ':'))) {
4015 name = htmlParseHTMLName_nonInvasive(ctxt);
4017 htmlParseErr(ctxt, XML_ERR_NAME_REQUIRED,
4018 "htmlParseStartTag: invalid element name\n",
4020 /* Dump the bogus tag like browsers do */
4021 while ((IS_CHAR_CH(CUR)) && (CUR != '>'))
4024 if (currentNode != NULL)
4025 xmlFree(currentNode);
4029 if (ctxt->name != NULL) {
4030 if (htmlCheckAutoClose(name, ctxt->name) == 1) {
4031 htmlAutoClose(ctxt, name);
4038 * Has this node been popped out during parsing of
4041 if ((ctxt->nameNr > 0) && (depth >= ctxt->nameNr) &&
4042 (!xmlStrEqual(currentNode, ctxt->name)))
4044 if (currentNode != NULL) xmlFree(currentNode);
4048 if ((CUR != 0) && ((xmlStrEqual(currentNode, BAD_CAST"script")) ||
4049 (xmlStrEqual(currentNode, BAD_CAST"style")))) {
4051 * Handle SCRIPT/STYLE separately
4053 htmlParseScript(ctxt);
4056 * Sometimes DOCTYPE arrives in the middle of the document
4058 if ((CUR == '<') && (NXT(1) == '!') &&
4059 (UPP(2) == 'D') && (UPP(3) == 'O') &&
4060 (UPP(4) == 'C') && (UPP(5) == 'T') &&
4061 (UPP(6) == 'Y') && (UPP(7) == 'P') &&
4063 htmlParseErr(ctxt, XML_HTML_STRUCURE_ERROR,
4064 "Misplaced DOCTYPE declaration\n",
4065 BAD_CAST "DOCTYPE" , NULL);
4066 htmlParseDocTypeDecl(ctxt);
4070 * First case : a comment
4072 if ((CUR == '<') && (NXT(1) == '!') &&
4073 (NXT(2) == '-') && (NXT(3) == '-')) {
4074 htmlParseComment(ctxt);
4078 * Second case : a Processing Instruction.
4080 else if ((CUR == '<') && (NXT(1) == '?')) {
4085 * Third case : a sub-element.
4087 else if (CUR == '<') {
4088 htmlParseElement(ctxt);
4092 * Fourth case : a reference. If if has not been resolved,
4093 * parsing returns it's Name, create the node
4095 else if (CUR == '&') {
4096 htmlParseReference(ctxt);
4100 * Fifth case : end of the resource
4102 else if (CUR == 0) {
4103 htmlAutoCloseOnEnd(ctxt);
4108 * Last case, text. Note that References are handled directly.
4111 htmlParseCharData(ctxt);
4114 if (cons == ctxt->nbChars) {
4115 if (ctxt->node != NULL) {
4116 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
4117 "detected an error in element content\n",
4125 if (currentNode != NULL) xmlFree(currentNode);
4130 * @ctxt: an HTML parser context
4132 * parse an HTML element, this is highly recursive
4133 * this is kept for compatibility with previous code versions
4135 * [39] element ::= EmptyElemTag | STag content ETag
4137 * [41] Attribute ::= Name Eq AttValue
4141 htmlParseElement(htmlParserCtxtPtr ctxt) {
4142 const xmlChar *name;
4143 xmlChar *currentNode = NULL;
4144 const htmlElemDesc * info;
4145 htmlParserNodeInfo node_info;
4148 const xmlChar *oldptr;
4150 if ((ctxt == NULL) || (ctxt->input == NULL)) {
4151 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
4152 "htmlParseElement: context error\n", NULL, NULL);
4156 if (ctxt->instate == XML_PARSER_EOF)
4159 /* Capture start position */
4160 if (ctxt->record_info) {
4161 node_info.begin_pos = ctxt->input->consumed +
4162 (CUR_PTR - ctxt->input->base);
4163 node_info.begin_line = ctxt->input->line;
4166 failed = htmlParseStartTag(ctxt);
4168 if ((failed == -1) || (name == NULL)) {
4175 * Lookup the info for that element.
4177 info = htmlTagLookup(name);
4179 htmlParseErr(ctxt, XML_HTML_UNKNOWN_TAG,
4180 "Tag %s invalid\n", name, NULL);
4184 * Check for an Empty Element labeled the XML/SGML way
4186 if ((CUR == '/') && (NXT(1) == '>')) {
4188 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
4189 ctxt->sax->endElement(ctxt->userData, name);
4197 htmlParseErr(ctxt, XML_ERR_GT_REQUIRED,
4198 "Couldn't find end of Start Tag %s\n", name, NULL);
4201 * end of parsing of this node.
4203 if (xmlStrEqual(name, ctxt->name)) {
4209 * Capture end position and add node
4211 if (ctxt->record_info) {
4212 node_info.end_pos = ctxt->input->consumed +
4213 (CUR_PTR - ctxt->input->base);
4214 node_info.end_line = ctxt->input->line;
4215 node_info.node = ctxt->node;
4216 xmlParserAddNodeInfo(ctxt, &node_info);
4222 * Check for an Empty Element from DTD definition
4224 if ((info != NULL) && (info->empty)) {
4225 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
4226 ctxt->sax->endElement(ctxt->userData, name);
4232 * Parse the content of the element:
4234 currentNode = xmlStrdup(ctxt->name);
4235 depth = ctxt->nameNr;
4236 while (IS_CHAR_CH(CUR)) {
4237 oldptr = ctxt->input->cur;
4238 htmlParseContent(ctxt);
4239 if (oldptr==ctxt->input->cur) break;
4240 if (ctxt->nameNr < depth) break;
4244 * Capture end position and add node
4246 if ( currentNode != NULL && ctxt->record_info ) {
4247 node_info.end_pos = ctxt->input->consumed +
4248 (CUR_PTR - ctxt->input->base);
4249 node_info.end_line = ctxt->input->line;
4250 node_info.node = ctxt->node;
4251 xmlParserAddNodeInfo(ctxt, &node_info);
4253 if (!IS_CHAR_CH(CUR)) {
4254 htmlAutoCloseOnEnd(ctxt);
4257 if (currentNode != NULL)
4258 xmlFree(currentNode);
4262 htmlParserFinishElementParsing(htmlParserCtxtPtr ctxt) {
4264 * Capture end position and add node
4266 if ( ctxt->node != NULL && ctxt->record_info ) {
4267 ctxt->nodeInfo->end_pos = ctxt->input->consumed +
4268 (CUR_PTR - ctxt->input->base);
4269 ctxt->nodeInfo->end_line = ctxt->input->line;
4270 ctxt->nodeInfo->node = ctxt->node;
4271 xmlParserAddNodeInfo(ctxt, ctxt->nodeInfo);
4272 htmlNodeInfoPop(ctxt);
4274 if (!IS_CHAR_CH(CUR)) {
4275 htmlAutoCloseOnEnd(ctxt);
4280 * htmlParseElementInternal:
4281 * @ctxt: an HTML parser context
4283 * parse an HTML element, new version, non recursive
4285 * [39] element ::= EmptyElemTag | STag content ETag
4287 * [41] Attribute ::= Name Eq AttValue
4291 htmlParseElementInternal(htmlParserCtxtPtr ctxt) {
4292 const xmlChar *name;
4293 const htmlElemDesc * info;
4294 htmlParserNodeInfo node_info;
4297 if ((ctxt == NULL) || (ctxt->input == NULL)) {
4298 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
4299 "htmlParseElementInternal: context error\n", NULL, NULL);
4303 if (ctxt->instate == XML_PARSER_EOF)
4306 /* Capture start position */
4307 if (ctxt->record_info) {
4308 node_info.begin_pos = ctxt->input->consumed +
4309 (CUR_PTR - ctxt->input->base);
4310 node_info.begin_line = ctxt->input->line;
4313 failed = htmlParseStartTag(ctxt);
4315 if ((failed == -1) || (name == NULL)) {
4322 * Lookup the info for that element.
4324 info = htmlTagLookup(name);
4326 htmlParseErr(ctxt, XML_HTML_UNKNOWN_TAG,
4327 "Tag %s invalid\n", name, NULL);
4331 * Check for an Empty Element labeled the XML/SGML way
4333 if ((CUR == '/') && (NXT(1) == '>')) {
4335 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
4336 ctxt->sax->endElement(ctxt->userData, name);
4344 htmlParseErr(ctxt, XML_ERR_GT_REQUIRED,
4345 "Couldn't find end of Start Tag %s\n", name, NULL);
4348 * end of parsing of this node.
4350 if (xmlStrEqual(name, ctxt->name)) {
4355 if (ctxt->record_info)
4356 htmlNodeInfoPush(ctxt, &node_info);
4357 htmlParserFinishElementParsing(ctxt);
4362 * Check for an Empty Element from DTD definition
4364 if ((info != NULL) && (info->empty)) {
4365 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
4366 ctxt->sax->endElement(ctxt->userData, name);
4371 if (ctxt->record_info)
4372 htmlNodeInfoPush(ctxt, &node_info);
4376 * htmlParseContentInternal:
4377 * @ctxt: an HTML parser context
4379 * Parse a content: comment, sub-element, reference or text.
4380 * New version for non recursive htmlParseElementInternal
4384 htmlParseContentInternal(htmlParserCtxtPtr ctxt) {
4385 xmlChar *currentNode;
4387 const xmlChar *name;
4389 currentNode = xmlStrdup(ctxt->name);
4390 depth = ctxt->nameNr;
4392 long cons = ctxt->nbChars;
4396 if (ctxt->instate == XML_PARSER_EOF)
4400 * Our tag or one of it's parent or children is ending.
4402 if ((CUR == '<') && (NXT(1) == '/')) {
4403 if (htmlParseEndTag(ctxt) &&
4404 ((currentNode != NULL) || (ctxt->nameNr == 0))) {
4405 if (currentNode != NULL)
4406 xmlFree(currentNode);
4408 currentNode = xmlStrdup(ctxt->name);
4409 depth = ctxt->nameNr;
4411 continue; /* while */
4414 else if ((CUR == '<') &&
4415 ((IS_ASCII_LETTER(NXT(1))) ||
4416 (NXT(1) == '_') || (NXT(1) == ':'))) {
4417 name = htmlParseHTMLName_nonInvasive(ctxt);
4419 htmlParseErr(ctxt, XML_ERR_NAME_REQUIRED,
4420 "htmlParseStartTag: invalid element name\n",
4422 /* Dump the bogus tag like browsers do */
4423 while ((IS_CHAR_CH(CUR)) && (CUR != '>'))
4426 htmlParserFinishElementParsing(ctxt);
4427 if (currentNode != NULL)
4428 xmlFree(currentNode);
4430 currentNode = xmlStrdup(ctxt->name);
4431 depth = ctxt->nameNr;
4435 if (ctxt->name != NULL) {
4436 if (htmlCheckAutoClose(name, ctxt->name) == 1) {
4437 htmlAutoClose(ctxt, name);
4444 * Has this node been popped out during parsing of
4447 if ((ctxt->nameNr > 0) && (depth >= ctxt->nameNr) &&
4448 (!xmlStrEqual(currentNode, ctxt->name)))
4450 htmlParserFinishElementParsing(ctxt);
4451 if (currentNode != NULL) xmlFree(currentNode);
4453 currentNode = xmlStrdup(ctxt->name);
4454 depth = ctxt->nameNr;
4458 if ((CUR != 0) && ((xmlStrEqual(currentNode, BAD_CAST"script")) ||
4459 (xmlStrEqual(currentNode, BAD_CAST"style")))) {
4461 * Handle SCRIPT/STYLE separately
4463 htmlParseScript(ctxt);
4466 * Sometimes DOCTYPE arrives in the middle of the document
4468 if ((CUR == '<') && (NXT(1) == '!') &&
4469 (UPP(2) == 'D') && (UPP(3) == 'O') &&
4470 (UPP(4) == 'C') && (UPP(5) == 'T') &&
4471 (UPP(6) == 'Y') && (UPP(7) == 'P') &&
4473 htmlParseErr(ctxt, XML_HTML_STRUCURE_ERROR,
4474 "Misplaced DOCTYPE declaration\n",
4475 BAD_CAST "DOCTYPE" , NULL);
4476 htmlParseDocTypeDecl(ctxt);
4480 * First case : a comment
4482 if ((CUR == '<') && (NXT(1) == '!') &&
4483 (NXT(2) == '-') && (NXT(3) == '-')) {
4484 htmlParseComment(ctxt);
4488 * Second case : a Processing Instruction.
4490 else if ((CUR == '<') && (NXT(1) == '?')) {
4495 * Third case : a sub-element.
4497 else if (CUR == '<') {
4498 htmlParseElementInternal(ctxt);
4499 if (currentNode != NULL) xmlFree(currentNode);
4501 currentNode = xmlStrdup(ctxt->name);
4502 depth = ctxt->nameNr;
4506 * Fourth case : a reference. If if has not been resolved,
4507 * parsing returns it's Name, create the node
4509 else if (CUR == '&') {
4510 htmlParseReference(ctxt);
4514 * Fifth case : end of the resource
4516 else if (CUR == 0) {
4517 htmlAutoCloseOnEnd(ctxt);
4522 * Last case, text. Note that References are handled directly.
4525 htmlParseCharData(ctxt);
4528 if (cons == ctxt->nbChars) {
4529 if (ctxt->node != NULL) {
4530 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
4531 "detected an error in element content\n",
4539 if (currentNode != NULL) xmlFree(currentNode);
4544 * @ctxt: an HTML parser context
4546 * Parse a content: comment, sub-element, reference or text.
4547 * This is the entry point when called from parser.c
4551 __htmlParseContent(void *ctxt) {
4553 htmlParseContentInternal((htmlParserCtxtPtr) ctxt);
4557 * htmlParseDocument:
4558 * @ctxt: an HTML parser context
4560 * parse an HTML document (and build a tree if using the standard SAX
4563 * Returns 0, -1 in case of error. the parser context is augmented
4564 * as a result of the parsing.
4568 htmlParseDocument(htmlParserCtxtPtr ctxt) {
4570 xmlCharEncoding enc;
4575 htmlDefaultSAXHandlerInit();
4577 if ((ctxt == NULL) || (ctxt->input == NULL)) {
4578 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
4579 "htmlParseDocument: context error\n", NULL, NULL);
4580 return(XML_ERR_INTERNAL_ERROR);
4583 ctxt->linenumbers = 1;
4586 * SAX: beginning of the document processing.
4588 if ((ctxt->sax) && (ctxt->sax->setDocumentLocator))
4589 ctxt->sax->setDocumentLocator(ctxt->userData, &xmlDefaultSAXLocator);
4591 if ((ctxt->encoding == (const xmlChar *)XML_CHAR_ENCODING_NONE) &&
4592 ((ctxt->input->end - ctxt->input->cur) >= 4)) {
4594 * Get the 4 first bytes and decode the charset
4595 * if enc != XML_CHAR_ENCODING_NONE
4596 * plug some encoding conversion routines.
4602 enc = xmlDetectCharEncoding(&start[0], 4);
4603 if (enc != XML_CHAR_ENCODING_NONE) {
4604 xmlSwitchEncoding(ctxt, enc);
4609 * Wipe out everything which is before the first '<'
4613 htmlParseErr(ctxt, XML_ERR_DOCUMENT_EMPTY,
4614 "Document is empty\n", NULL, NULL);
4617 if ((ctxt->sax) && (ctxt->sax->startDocument) && (!ctxt->disableSAX))
4618 ctxt->sax->startDocument(ctxt->userData);
4622 * Parse possible comments and PIs before any content
4624 while (((CUR == '<') && (NXT(1) == '!') &&
4625 (NXT(2) == '-') && (NXT(3) == '-')) ||
4626 ((CUR == '<') && (NXT(1) == '?'))) {
4627 htmlParseComment(ctxt);
4634 * Then possibly doc type declaration(s) and more Misc
4635 * (doctypedecl Misc*)?
4637 if ((CUR == '<') && (NXT(1) == '!') &&
4638 (UPP(2) == 'D') && (UPP(3) == 'O') &&
4639 (UPP(4) == 'C') && (UPP(5) == 'T') &&
4640 (UPP(6) == 'Y') && (UPP(7) == 'P') &&
4642 htmlParseDocTypeDecl(ctxt);
4647 * Parse possible comments and PIs before any content
4649 while (((CUR == '<') && (NXT(1) == '!') &&
4650 (NXT(2) == '-') && (NXT(3) == '-')) ||
4651 ((CUR == '<') && (NXT(1) == '?'))) {
4652 htmlParseComment(ctxt);
4658 * Time to start parsing the tree itself
4660 htmlParseContentInternal(ctxt);
4666 htmlAutoCloseOnEnd(ctxt);
4670 * SAX: end of the document processing.
4672 if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
4673 ctxt->sax->endDocument(ctxt->userData);
4675 if ((!(ctxt->options & HTML_PARSE_NODEFDTD)) && (ctxt->myDoc != NULL)) {
4676 dtd = xmlGetIntSubset(ctxt->myDoc);
4678 ctxt->myDoc->intSubset =
4679 xmlCreateIntSubset(ctxt->myDoc, BAD_CAST "html",
4680 BAD_CAST "-//W3C//DTD HTML 4.0 Transitional//EN",
4681 BAD_CAST "http://www.w3.org/TR/REC-html40/loose.dtd");
4683 if (! ctxt->wellFormed) return(-1);
4688 /************************************************************************
4690 * Parser contexts handling *
4692 ************************************************************************/
4695 * htmlInitParserCtxt:
4696 * @ctxt: an HTML parser context
4698 * Initialize a parser context
4700 * Returns 0 in case of success and -1 in case of error
4704 htmlInitParserCtxt(htmlParserCtxtPtr ctxt)
4706 htmlSAXHandler *sax;
4708 if (ctxt == NULL) return(-1);
4709 memset(ctxt, 0, sizeof(htmlParserCtxt));
4711 ctxt->dict = xmlDictCreate();
4712 if (ctxt->dict == NULL) {
4713 htmlErrMemory(NULL, "htmlInitParserCtxt: out of memory\n");
4716 sax = (htmlSAXHandler *) xmlMalloc(sizeof(htmlSAXHandler));
4718 htmlErrMemory(NULL, "htmlInitParserCtxt: out of memory\n");
4722 memset(sax, 0, sizeof(htmlSAXHandler));
4724 /* Allocate the Input stack */
4725 ctxt->inputTab = (htmlParserInputPtr *)
4726 xmlMalloc(5 * sizeof(htmlParserInputPtr));
4727 if (ctxt->inputTab == NULL) {
4728 htmlErrMemory(NULL, "htmlInitParserCtxt: out of memory\n");
4737 ctxt->version = NULL;
4738 ctxt->encoding = NULL;
4739 ctxt->standalone = -1;
4740 ctxt->instate = XML_PARSER_START;
4742 /* Allocate the Node stack */
4743 ctxt->nodeTab = (htmlNodePtr *) xmlMalloc(10 * sizeof(htmlNodePtr));
4744 if (ctxt->nodeTab == NULL) {
4745 htmlErrMemory(NULL, "htmlInitParserCtxt: out of memory\n");
4758 /* Allocate the Name stack */
4759 ctxt->nameTab = (const xmlChar **) xmlMalloc(10 * sizeof(xmlChar *));
4760 if (ctxt->nameTab == NULL) {
4761 htmlErrMemory(NULL, "htmlInitParserCtxt: out of memory\n");
4777 ctxt->nodeInfoTab = NULL;
4778 ctxt->nodeInfoNr = 0;
4779 ctxt->nodeInfoMax = 0;
4781 if (sax == NULL) ctxt->sax = (xmlSAXHandlerPtr) &htmlDefaultSAXHandler;
4784 memcpy(sax, &htmlDefaultSAXHandler, sizeof(xmlSAXHandlerV1));
4786 ctxt->userData = ctxt;
4788 ctxt->wellFormed = 1;
4789 ctxt->replaceEntities = 0;
4790 ctxt->linenumbers = xmlLineNumbersDefaultValue;
4792 ctxt->vctxt.finishDtd = XML_CTXT_FINISH_DTD_0;
4793 ctxt->vctxt.userData = ctxt;
4794 ctxt->vctxt.error = xmlParserValidityError;
4795 ctxt->vctxt.warning = xmlParserValidityWarning;
4796 ctxt->record_info = 0;
4799 ctxt->checkIndex = 0;
4800 ctxt->catalogs = NULL;
4801 xmlInitNodeInfoSeq(&ctxt->node_seq);
4806 * htmlFreeParserCtxt:
4807 * @ctxt: an HTML parser context
4809 * Free all the memory used by a parser context. However the parsed
4810 * document in ctxt->myDoc is not freed.
4814 htmlFreeParserCtxt(htmlParserCtxtPtr ctxt)
4816 xmlFreeParserCtxt(ctxt);
4820 * htmlNewParserCtxt:
4822 * Allocate and initialize a new parser context.
4824 * Returns the htmlParserCtxtPtr or NULL in case of allocation error
4828 htmlNewParserCtxt(void)
4830 xmlParserCtxtPtr ctxt;
4832 ctxt = (xmlParserCtxtPtr) xmlMalloc(sizeof(xmlParserCtxt));
4834 htmlErrMemory(NULL, "NewParserCtxt: out of memory\n");
4837 memset(ctxt, 0, sizeof(xmlParserCtxt));
4838 if (htmlInitParserCtxt(ctxt) < 0) {
4839 htmlFreeParserCtxt(ctxt);
4846 * htmlCreateMemoryParserCtxt:
4847 * @buffer: a pointer to a char array
4848 * @size: the size of the array
4850 * Create a parser context for an HTML in-memory document.
4852 * Returns the new parser context or NULL
4855 htmlCreateMemoryParserCtxt(const char *buffer, int size) {
4856 xmlParserCtxtPtr ctxt;
4857 xmlParserInputPtr input;
4858 xmlParserInputBufferPtr buf;
4865 ctxt = htmlNewParserCtxt();
4869 buf = xmlParserInputBufferCreateMem(buffer, size, XML_CHAR_ENCODING_NONE);
4870 if (buf == NULL) return(NULL);
4872 input = xmlNewInputStream(ctxt);
4873 if (input == NULL) {
4874 xmlFreeParserCtxt(ctxt);
4878 input->filename = NULL;
4880 input->base = input->buf->buffer->content;
4881 input->cur = input->buf->buffer->content;
4882 input->end = &input->buf->buffer->content[input->buf->buffer->use];
4884 inputPush(ctxt, input);
4889 * htmlCreateDocParserCtxt:
4890 * @cur: a pointer to an array of xmlChar
4891 * @encoding: a free form C string describing the HTML document encoding, or NULL
4893 * Create a parser context for an HTML document.
4895 * TODO: check the need to add encoding handling there
4897 * Returns the new parser context or NULL
4899 static htmlParserCtxtPtr
4900 htmlCreateDocParserCtxt(const xmlChar *cur, const char *encoding) {
4902 htmlParserCtxtPtr ctxt;
4906 len = xmlStrlen(cur);
4907 ctxt = htmlCreateMemoryParserCtxt((char *)cur, len);
4911 if (encoding != NULL) {
4912 xmlCharEncoding enc;
4913 xmlCharEncodingHandlerPtr handler;
4915 if (ctxt->input->encoding != NULL)
4916 xmlFree((xmlChar *) ctxt->input->encoding);
4917 ctxt->input->encoding = xmlStrdup((const xmlChar *) encoding);
4919 enc = xmlParseCharEncoding(encoding);
4921 * registered set of known encodings
4923 if (enc != XML_CHAR_ENCODING_ERROR) {
4924 xmlSwitchEncoding(ctxt, enc);
4925 if (ctxt->errNo == XML_ERR_UNSUPPORTED_ENCODING) {
4926 htmlParseErr(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
4927 "Unsupported encoding %s\n",
4928 (const xmlChar *) encoding, NULL);
4932 * fallback for unknown encodings
4934 handler = xmlFindCharEncodingHandler((const char *) encoding);
4935 if (handler != NULL) {
4936 xmlSwitchToEncoding(ctxt, handler);
4938 htmlParseErr(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
4939 "Unsupported encoding %s\n",
4940 (const xmlChar *) encoding, NULL);
4947 #ifdef LIBXML_PUSH_ENABLED
4948 /************************************************************************
4950 * Progressive parsing interfaces *
4952 ************************************************************************/
4955 * htmlParseLookupSequence:
4956 * @ctxt: an HTML parser context
4957 * @first: the first char to lookup
4958 * @next: the next char to lookup or zero
4959 * @third: the next char to lookup or zero
4960 * @comment: flag to force checking inside comments
4962 * Try to find if a sequence (first, next, third) or just (first next) or
4963 * (first) is available in the input stream.
4964 * This function has a side effect of (possibly) incrementing ctxt->checkIndex
4965 * to avoid rescanning sequences of bytes, it DOES change the state of the
4966 * parser, do not use liberally.
4967 * This is basically similar to xmlParseLookupSequence()
4969 * Returns the index to the current parsing point if the full sequence
4970 * is available, -1 otherwise.
4973 htmlParseLookupSequence(htmlParserCtxtPtr ctxt, xmlChar first,
4974 xmlChar next, xmlChar third, int iscomment,
4978 htmlParserInputPtr in;
4982 char valdellim = 0x0;
4988 base = in->cur - in->base;
4992 if (ctxt->checkIndex > base)
4993 base = ctxt->checkIndex;
4995 if (in->buf == NULL) {
4999 buf = in->buf->buffer->content;
5000 len = in->buf->buffer->use;
5003 /* take into account the sequence length */
5008 for (; base < len; base++) {
5009 if ((!incomment) && (base + 4 < len) && (!iscomment)) {
5010 if ((buf[base] == '<') && (buf[base + 1] == '!') &&
5011 (buf[base + 2] == '-') && (buf[base + 3] == '-')) {
5013 /* do not increment past <! - some people use <!--> */
5017 if (ignoreattrval) {
5018 if (buf[base] == '"' || buf[base] == '\'') {
5020 if (buf[base] == valdellim) {
5025 valdellim = buf[base];
5029 } else if (invalue) {
5036 if ((buf[base] == '-') && (buf[base + 1] == '-') &&
5037 (buf[base + 2] == '>')) {
5043 if (buf[base] == first) {
5045 if ((buf[base + 1] != next) || (buf[base + 2] != third))
5047 } else if (next != 0) {
5048 if (buf[base + 1] != next)
5051 ctxt->checkIndex = 0;
5054 xmlGenericError(xmlGenericErrorContext,
5055 "HPP: lookup '%c' found at %d\n",
5057 else if (third == 0)
5058 xmlGenericError(xmlGenericErrorContext,
5059 "HPP: lookup '%c%c' found at %d\n",
5062 xmlGenericError(xmlGenericErrorContext,
5063 "HPP: lookup '%c%c%c' found at %d\n",
5064 first, next, third, base);
5066 return (base - (in->cur - in->base));
5069 if ((!incomment) && (!invalue))
5070 ctxt->checkIndex = base;
5073 xmlGenericError(xmlGenericErrorContext,
5074 "HPP: lookup '%c' failed\n", first);
5075 else if (third == 0)
5076 xmlGenericError(xmlGenericErrorContext,
5077 "HPP: lookup '%c%c' failed\n", first, next);
5079 xmlGenericError(xmlGenericErrorContext,
5080 "HPP: lookup '%c%c%c' failed\n", first, next,
5087 * htmlParseLookupChars:
5088 * @ctxt: an HTML parser context
5089 * @stop: Array of chars, which stop the lookup.
5090 * @stopLen: Length of stop-Array
5092 * Try to find if any char of the stop-Array is available in the input
5094 * This function has a side effect of (possibly) incrementing ctxt->checkIndex
5095 * to avoid rescanning sequences of bytes, it DOES change the state of the
5096 * parser, do not use liberally.
5098 * Returns the index to the current parsing point if a stopChar
5099 * is available, -1 otherwise.
5102 htmlParseLookupChars(htmlParserCtxtPtr ctxt, const xmlChar * stop,
5106 htmlParserInputPtr in;
5115 base = in->cur - in->base;
5119 if (ctxt->checkIndex > base)
5120 base = ctxt->checkIndex;
5122 if (in->buf == NULL) {
5126 buf = in->buf->buffer->content;
5127 len = in->buf->buffer->use;
5130 for (; base < len; base++) {
5131 if (!incomment && (base + 4 < len)) {
5132 if ((buf[base] == '<') && (buf[base + 1] == '!') &&
5133 (buf[base + 2] == '-') && (buf[base + 3] == '-')) {
5135 /* do not increment past <! - some people use <!--> */
5142 if ((buf[base] == '-') && (buf[base + 1] == '-') &&
5143 (buf[base + 2] == '>')) {
5149 for (i = 0; i < stopLen; ++i) {
5150 if (buf[base] == stop[i]) {
5151 ctxt->checkIndex = 0;
5152 return (base - (in->cur - in->base));
5156 ctxt->checkIndex = base;
5161 * htmlParseTryOrFinish:
5162 * @ctxt: an HTML parser context
5163 * @terminate: last chunk indicator
5165 * Try to progress on parsing
5167 * Returns zero if no parsing was possible
5170 htmlParseTryOrFinish(htmlParserCtxtPtr ctxt, int terminate) {
5172 htmlParserInputPtr in;
5177 switch (ctxt->instate) {
5178 case XML_PARSER_EOF:
5179 xmlGenericError(xmlGenericErrorContext,
5180 "HPP: try EOF\n"); break;
5181 case XML_PARSER_START:
5182 xmlGenericError(xmlGenericErrorContext,
5183 "HPP: try START\n"); break;
5184 case XML_PARSER_MISC:
5185 xmlGenericError(xmlGenericErrorContext,
5186 "HPP: try MISC\n");break;
5187 case XML_PARSER_COMMENT:
5188 xmlGenericError(xmlGenericErrorContext,
5189 "HPP: try COMMENT\n");break;
5190 case XML_PARSER_PROLOG:
5191 xmlGenericError(xmlGenericErrorContext,
5192 "HPP: try PROLOG\n");break;
5193 case XML_PARSER_START_TAG:
5194 xmlGenericError(xmlGenericErrorContext,
5195 "HPP: try START_TAG\n");break;
5196 case XML_PARSER_CONTENT:
5197 xmlGenericError(xmlGenericErrorContext,
5198 "HPP: try CONTENT\n");break;
5199 case XML_PARSER_CDATA_SECTION:
5200 xmlGenericError(xmlGenericErrorContext,
5201 "HPP: try CDATA_SECTION\n");break;
5202 case XML_PARSER_END_TAG:
5203 xmlGenericError(xmlGenericErrorContext,
5204 "HPP: try END_TAG\n");break;
5205 case XML_PARSER_ENTITY_DECL:
5206 xmlGenericError(xmlGenericErrorContext,
5207 "HPP: try ENTITY_DECL\n");break;
5208 case XML_PARSER_ENTITY_VALUE:
5209 xmlGenericError(xmlGenericErrorContext,
5210 "HPP: try ENTITY_VALUE\n");break;
5211 case XML_PARSER_ATTRIBUTE_VALUE:
5212 xmlGenericError(xmlGenericErrorContext,
5213 "HPP: try ATTRIBUTE_VALUE\n");break;
5214 case XML_PARSER_DTD:
5215 xmlGenericError(xmlGenericErrorContext,
5216 "HPP: try DTD\n");break;
5217 case XML_PARSER_EPILOG:
5218 xmlGenericError(xmlGenericErrorContext,
5219 "HPP: try EPILOG\n");break;
5221 xmlGenericError(xmlGenericErrorContext,
5222 "HPP: try PI\n");break;
5223 case XML_PARSER_SYSTEM_LITERAL:
5224 xmlGenericError(xmlGenericErrorContext,
5225 "HPP: try SYSTEM_LITERAL\n");break;
5232 if (in == NULL) break;
5233 if (in->buf == NULL)
5234 avail = in->length - (in->cur - in->base);
5236 avail = in->buf->buffer->use - (in->cur - in->base);
5237 if ((avail == 0) && (terminate)) {
5238 htmlAutoCloseOnEnd(ctxt);
5239 if ((ctxt->nameNr == 0) && (ctxt->instate != XML_PARSER_EOF)) {
5241 * SAX: end of the document processing.
5243 ctxt->instate = XML_PARSER_EOF;
5244 if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
5245 ctxt->sax->endDocument(ctxt->userData);
5256 switch (ctxt->instate) {
5257 case XML_PARSER_EOF:
5259 * Document parsing is done !
5262 case XML_PARSER_START:
5264 * Very first chars read from the document flow.
5267 if (IS_BLANK_CH(cur)) {
5269 if (in->buf == NULL)
5270 avail = in->length - (in->cur - in->base);
5272 avail = in->buf->buffer->use - (in->cur - in->base);
5274 if ((ctxt->sax) && (ctxt->sax->setDocumentLocator))
5275 ctxt->sax->setDocumentLocator(ctxt->userData,
5276 &xmlDefaultSAXLocator);
5277 if ((ctxt->sax) && (ctxt->sax->startDocument) &&
5278 (!ctxt->disableSAX))
5279 ctxt->sax->startDocument(ctxt->userData);
5283 if ((cur == '<') && (next == '!') &&
5284 (UPP(2) == 'D') && (UPP(3) == 'O') &&
5285 (UPP(4) == 'C') && (UPP(5) == 'T') &&
5286 (UPP(6) == 'Y') && (UPP(7) == 'P') &&
5289 (htmlParseLookupSequence(ctxt, '>', 0, 0, 0, 1) < 0))
5292 xmlGenericError(xmlGenericErrorContext,
5293 "HPP: Parsing internal subset\n");
5295 htmlParseDocTypeDecl(ctxt);
5296 ctxt->instate = XML_PARSER_PROLOG;
5298 xmlGenericError(xmlGenericErrorContext,
5299 "HPP: entering PROLOG\n");
5302 ctxt->instate = XML_PARSER_MISC;
5304 xmlGenericError(xmlGenericErrorContext,
5305 "HPP: entering MISC\n");
5309 case XML_PARSER_MISC:
5311 if (in->buf == NULL)
5312 avail = in->length - (in->cur - in->base);
5314 avail = in->buf->buffer->use - (in->cur - in->base);
5319 if ((cur == '<') && (next == '!') &&
5320 (in->cur[2] == '-') && (in->cur[3] == '-')) {
5322 (htmlParseLookupSequence(ctxt, '-', '-', '>', 1, 1) < 0))
5325 xmlGenericError(xmlGenericErrorContext,
5326 "HPP: Parsing Comment\n");
5328 htmlParseComment(ctxt);
5329 ctxt->instate = XML_PARSER_MISC;
5330 } else if ((cur == '<') && (next == '?')) {
5332 (htmlParseLookupSequence(ctxt, '>', 0, 0, 0, 1) < 0))
5335 xmlGenericError(xmlGenericErrorContext,
5336 "HPP: Parsing PI\n");
5339 ctxt->instate = XML_PARSER_MISC;
5340 } else if ((cur == '<') && (next == '!') &&
5341 (UPP(2) == 'D') && (UPP(3) == 'O') &&
5342 (UPP(4) == 'C') && (UPP(5) == 'T') &&
5343 (UPP(6) == 'Y') && (UPP(7) == 'P') &&
5346 (htmlParseLookupSequence(ctxt, '>', 0, 0, 0, 1) < 0))
5349 xmlGenericError(xmlGenericErrorContext,
5350 "HPP: Parsing internal subset\n");
5352 htmlParseDocTypeDecl(ctxt);
5353 ctxt->instate = XML_PARSER_PROLOG;
5355 xmlGenericError(xmlGenericErrorContext,
5356 "HPP: entering PROLOG\n");
5358 } else if ((cur == '<') && (next == '!') &&
5362 ctxt->instate = XML_PARSER_START_TAG;
5364 xmlGenericError(xmlGenericErrorContext,
5365 "HPP: entering START_TAG\n");
5369 case XML_PARSER_PROLOG:
5371 if (in->buf == NULL)
5372 avail = in->length - (in->cur - in->base);
5374 avail = in->buf->buffer->use - (in->cur - in->base);
5379 if ((cur == '<') && (next == '!') &&
5380 (in->cur[2] == '-') && (in->cur[3] == '-')) {
5382 (htmlParseLookupSequence(ctxt, '-', '-', '>', 1, 1) < 0))
5385 xmlGenericError(xmlGenericErrorContext,
5386 "HPP: Parsing Comment\n");
5388 htmlParseComment(ctxt);
5389 ctxt->instate = XML_PARSER_PROLOG;
5390 } else if ((cur == '<') && (next == '?')) {
5392 (htmlParseLookupSequence(ctxt, '>', 0, 0, 0, 1) < 0))
5395 xmlGenericError(xmlGenericErrorContext,
5396 "HPP: Parsing PI\n");
5399 ctxt->instate = XML_PARSER_PROLOG;
5400 } else if ((cur == '<') && (next == '!') &&
5404 ctxt->instate = XML_PARSER_START_TAG;
5406 xmlGenericError(xmlGenericErrorContext,
5407 "HPP: entering START_TAG\n");
5411 case XML_PARSER_EPILOG:
5412 if (in->buf == NULL)
5413 avail = in->length - (in->cur - in->base);
5415 avail = in->buf->buffer->use - (in->cur - in->base);
5419 if (IS_BLANK_CH(cur)) {
5420 htmlParseCharData(ctxt);
5426 if ((cur == '<') && (next == '!') &&
5427 (in->cur[2] == '-') && (in->cur[3] == '-')) {
5429 (htmlParseLookupSequence(ctxt, '-', '-', '>', 1, 1) < 0))
5432 xmlGenericError(xmlGenericErrorContext,
5433 "HPP: Parsing Comment\n");
5435 htmlParseComment(ctxt);
5436 ctxt->instate = XML_PARSER_EPILOG;
5437 } else if ((cur == '<') && (next == '?')) {
5439 (htmlParseLookupSequence(ctxt, '>', 0, 0, 0, 1) < 0))
5442 xmlGenericError(xmlGenericErrorContext,
5443 "HPP: Parsing PI\n");
5446 ctxt->instate = XML_PARSER_EPILOG;
5447 } else if ((cur == '<') && (next == '!') &&
5451 ctxt->errNo = XML_ERR_DOCUMENT_END;
5452 ctxt->wellFormed = 0;
5453 ctxt->instate = XML_PARSER_EOF;
5455 xmlGenericError(xmlGenericErrorContext,
5456 "HPP: entering EOF\n");
5458 if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
5459 ctxt->sax->endDocument(ctxt->userData);
5463 case XML_PARSER_START_TAG: {
5464 const xmlChar *name;
5466 const htmlElemDesc * info;
5472 ctxt->instate = XML_PARSER_CONTENT;
5474 xmlGenericError(xmlGenericErrorContext,
5475 "HPP: entering CONTENT\n");
5479 if (in->cur[1] == '/') {
5480 ctxt->instate = XML_PARSER_END_TAG;
5481 ctxt->checkIndex = 0;
5483 xmlGenericError(xmlGenericErrorContext,
5484 "HPP: entering END_TAG\n");
5489 (htmlParseLookupSequence(ctxt, '>', 0, 0, 0, 1) < 0))
5492 failed = htmlParseStartTag(ctxt);
5494 if ((failed == -1) ||
5502 * Lookup the info for that element.
5504 info = htmlTagLookup(name);
5506 htmlParseErr(ctxt, XML_HTML_UNKNOWN_TAG,
5507 "Tag %s invalid\n", name, NULL);
5511 * Check for an Empty Element labeled the XML/SGML way
5513 if ((CUR == '/') && (NXT(1) == '>')) {
5515 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
5516 ctxt->sax->endElement(ctxt->userData, name);
5518 ctxt->instate = XML_PARSER_CONTENT;
5520 xmlGenericError(xmlGenericErrorContext,
5521 "HPP: entering CONTENT\n");
5529 htmlParseErr(ctxt, XML_ERR_GT_REQUIRED,
5530 "Couldn't find end of Start Tag %s\n",
5534 * end of parsing of this node.
5536 if (xmlStrEqual(name, ctxt->name)) {
5541 ctxt->instate = XML_PARSER_CONTENT;
5543 xmlGenericError(xmlGenericErrorContext,
5544 "HPP: entering CONTENT\n");
5550 * Check for an Empty Element from DTD definition
5552 if ((info != NULL) && (info->empty)) {
5553 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
5554 ctxt->sax->endElement(ctxt->userData, name);
5557 ctxt->instate = XML_PARSER_CONTENT;
5559 xmlGenericError(xmlGenericErrorContext,
5560 "HPP: entering CONTENT\n");
5564 case XML_PARSER_CONTENT: {
5567 * Handle preparsed entities and charRef
5569 if (ctxt->token != 0) {
5570 xmlChar chr[2] = { 0 , 0 } ;
5572 chr[0] = (xmlChar) ctxt->token;
5573 htmlCheckParagraph(ctxt);
5574 if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
5575 ctxt->sax->characters(ctxt->userData, chr, 1);
5577 ctxt->checkIndex = 0;
5579 if ((avail == 1) && (terminate)) {
5581 if ((cur != '<') && (cur != '&')) {
5582 if (ctxt->sax != NULL) {
5583 if (IS_BLANK_CH(cur)) {
5584 if (ctxt->sax->ignorableWhitespace != NULL)
5585 ctxt->sax->ignorableWhitespace(
5586 ctxt->userData, &cur, 1);
5588 htmlCheckParagraph(ctxt);
5589 if (ctxt->sax->characters != NULL)
5590 ctxt->sax->characters(
5591 ctxt->userData, &cur, 1);
5595 ctxt->checkIndex = 0;
5604 cons = ctxt->nbChars;
5605 if ((xmlStrEqual(ctxt->name, BAD_CAST"script")) ||
5606 (xmlStrEqual(ctxt->name, BAD_CAST"style"))) {
5608 * Handle SCRIPT/STYLE separately
5614 idx = htmlParseLookupSequence(ctxt, '<', '/', 0, 0, 0);
5617 val = in->cur[idx + 2];
5618 if (val == 0) /* bad cut of input */
5621 htmlParseScript(ctxt);
5622 if ((cur == '<') && (next == '/')) {
5623 ctxt->instate = XML_PARSER_END_TAG;
5624 ctxt->checkIndex = 0;
5626 xmlGenericError(xmlGenericErrorContext,
5627 "HPP: entering END_TAG\n");
5633 * Sometimes DOCTYPE arrives in the middle of the document
5635 if ((cur == '<') && (next == '!') &&
5636 (UPP(2) == 'D') && (UPP(3) == 'O') &&
5637 (UPP(4) == 'C') && (UPP(5) == 'T') &&
5638 (UPP(6) == 'Y') && (UPP(7) == 'P') &&
5641 (htmlParseLookupSequence(ctxt, '>', 0, 0, 0, 1) < 0))
5643 htmlParseErr(ctxt, XML_HTML_STRUCURE_ERROR,
5644 "Misplaced DOCTYPE declaration\n",
5645 BAD_CAST "DOCTYPE" , NULL);
5646 htmlParseDocTypeDecl(ctxt);
5647 } else if ((cur == '<') && (next == '!') &&
5648 (in->cur[2] == '-') && (in->cur[3] == '-')) {
5650 (htmlParseLookupSequence(
5651 ctxt, '-', '-', '>', 1, 1) < 0))
5654 xmlGenericError(xmlGenericErrorContext,
5655 "HPP: Parsing Comment\n");
5657 htmlParseComment(ctxt);
5658 ctxt->instate = XML_PARSER_CONTENT;
5659 } else if ((cur == '<') && (next == '?')) {
5661 (htmlParseLookupSequence(ctxt, '>', 0, 0, 0, 1) < 0))
5664 xmlGenericError(xmlGenericErrorContext,
5665 "HPP: Parsing PI\n");
5668 ctxt->instate = XML_PARSER_CONTENT;
5669 } else if ((cur == '<') && (next == '!') && (avail < 4)) {
5671 } else if ((cur == '<') && (next == '/')) {
5672 ctxt->instate = XML_PARSER_END_TAG;
5673 ctxt->checkIndex = 0;
5675 xmlGenericError(xmlGenericErrorContext,
5676 "HPP: entering END_TAG\n");
5679 } else if (cur == '<') {
5680 ctxt->instate = XML_PARSER_START_TAG;
5681 ctxt->checkIndex = 0;
5683 xmlGenericError(xmlGenericErrorContext,
5684 "HPP: entering START_TAG\n");
5687 } else if (cur == '&') {
5689 (htmlParseLookupChars(ctxt,
5690 BAD_CAST "; >/", 4) < 0))
5693 xmlGenericError(xmlGenericErrorContext,
5694 "HPP: Parsing Reference\n");
5696 /* TODO: check generation of subtrees if noent !!! */
5697 htmlParseReference(ctxt);
5700 * check that the text sequence is complete
5701 * before handing out the data to the parser
5702 * to avoid problems with erroneous end of
5706 (htmlParseLookupChars(ctxt, BAD_CAST "<&", 2) < 0))
5708 ctxt->checkIndex = 0;
5710 xmlGenericError(xmlGenericErrorContext,
5711 "HPP: Parsing char data\n");
5713 htmlParseCharData(ctxt);
5716 if (cons == ctxt->nbChars) {
5717 if (ctxt->node != NULL) {
5718 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
5719 "detected an error in element content\n",
5728 case XML_PARSER_END_TAG:
5732 (htmlParseLookupSequence(ctxt, '>', 0, 0, 0, 1) < 0))
5734 htmlParseEndTag(ctxt);
5735 if (ctxt->nameNr == 0) {
5736 ctxt->instate = XML_PARSER_EPILOG;
5738 ctxt->instate = XML_PARSER_CONTENT;
5740 ctxt->checkIndex = 0;
5742 xmlGenericError(xmlGenericErrorContext,
5743 "HPP: entering CONTENT\n");
5746 case XML_PARSER_CDATA_SECTION:
5747 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
5748 "HPP: internal error, state == CDATA\n",
5750 ctxt->instate = XML_PARSER_CONTENT;
5751 ctxt->checkIndex = 0;
5753 xmlGenericError(xmlGenericErrorContext,
5754 "HPP: entering CONTENT\n");
5757 case XML_PARSER_DTD:
5758 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
5759 "HPP: internal error, state == DTD\n",
5761 ctxt->instate = XML_PARSER_CONTENT;
5762 ctxt->checkIndex = 0;
5764 xmlGenericError(xmlGenericErrorContext,
5765 "HPP: entering CONTENT\n");
5768 case XML_PARSER_COMMENT:
5769 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
5770 "HPP: internal error, state == COMMENT\n",
5772 ctxt->instate = XML_PARSER_CONTENT;
5773 ctxt->checkIndex = 0;
5775 xmlGenericError(xmlGenericErrorContext,
5776 "HPP: entering CONTENT\n");
5780 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
5781 "HPP: internal error, state == PI\n",
5783 ctxt->instate = XML_PARSER_CONTENT;
5784 ctxt->checkIndex = 0;
5786 xmlGenericError(xmlGenericErrorContext,
5787 "HPP: entering CONTENT\n");
5790 case XML_PARSER_ENTITY_DECL:
5791 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
5792 "HPP: internal error, state == ENTITY_DECL\n",
5794 ctxt->instate = XML_PARSER_CONTENT;
5795 ctxt->checkIndex = 0;
5797 xmlGenericError(xmlGenericErrorContext,
5798 "HPP: entering CONTENT\n");
5801 case XML_PARSER_ENTITY_VALUE:
5802 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
5803 "HPP: internal error, state == ENTITY_VALUE\n",
5805 ctxt->instate = XML_PARSER_CONTENT;
5806 ctxt->checkIndex = 0;
5808 xmlGenericError(xmlGenericErrorContext,
5809 "HPP: entering DTD\n");
5812 case XML_PARSER_ATTRIBUTE_VALUE:
5813 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
5814 "HPP: internal error, state == ATTRIBUTE_VALUE\n",
5816 ctxt->instate = XML_PARSER_START_TAG;
5817 ctxt->checkIndex = 0;
5819 xmlGenericError(xmlGenericErrorContext,
5820 "HPP: entering START_TAG\n");
5823 case XML_PARSER_SYSTEM_LITERAL:
5824 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
5825 "HPP: internal error, state == XML_PARSER_SYSTEM_LITERAL\n",
5827 ctxt->instate = XML_PARSER_CONTENT;
5828 ctxt->checkIndex = 0;
5830 xmlGenericError(xmlGenericErrorContext,
5831 "HPP: entering CONTENT\n");
5834 case XML_PARSER_IGNORE:
5835 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
5836 "HPP: internal error, state == XML_PARSER_IGNORE\n",
5838 ctxt->instate = XML_PARSER_CONTENT;
5839 ctxt->checkIndex = 0;
5841 xmlGenericError(xmlGenericErrorContext,
5842 "HPP: entering CONTENT\n");
5845 case XML_PARSER_PUBLIC_LITERAL:
5846 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
5847 "HPP: internal error, state == XML_PARSER_LITERAL\n",
5849 ctxt->instate = XML_PARSER_CONTENT;
5850 ctxt->checkIndex = 0;
5852 xmlGenericError(xmlGenericErrorContext,
5853 "HPP: entering CONTENT\n");
5860 if ((avail == 0) && (terminate)) {
5861 htmlAutoCloseOnEnd(ctxt);
5862 if ((ctxt->nameNr == 0) && (ctxt->instate != XML_PARSER_EOF)) {
5864 * SAX: end of the document processing.
5866 ctxt->instate = XML_PARSER_EOF;
5867 if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
5868 ctxt->sax->endDocument(ctxt->userData);
5871 if ((ctxt->myDoc != NULL) &&
5872 ((terminate) || (ctxt->instate == XML_PARSER_EOF) ||
5873 (ctxt->instate == XML_PARSER_EPILOG))) {
5875 dtd = xmlGetIntSubset(ctxt->myDoc);
5877 ctxt->myDoc->intSubset =
5878 xmlCreateIntSubset(ctxt->myDoc, BAD_CAST "html",
5879 BAD_CAST "-//W3C//DTD HTML 4.0 Transitional//EN",
5880 BAD_CAST "http://www.w3.org/TR/REC-html40/loose.dtd");
5883 xmlGenericError(xmlGenericErrorContext, "HPP: done %d\n", ret);
5890 * @ctxt: an HTML parser context
5891 * @chunk: an char array
5892 * @size: the size in byte of the chunk
5893 * @terminate: last chunk indicator
5895 * Parse a Chunk of memory
5897 * Returns zero if no error, the xmlParserErrors otherwise.
5900 htmlParseChunk(htmlParserCtxtPtr ctxt, const char *chunk, int size,
5902 if ((ctxt == NULL) || (ctxt->input == NULL)) {
5903 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
5904 "htmlParseChunk: context error\n", NULL, NULL);
5905 return(XML_ERR_INTERNAL_ERROR);
5907 if ((size > 0) && (chunk != NULL) && (ctxt->input != NULL) &&
5908 (ctxt->input->buf != NULL) && (ctxt->instate != XML_PARSER_EOF)) {
5909 int base = ctxt->input->base - ctxt->input->buf->buffer->content;
5910 int cur = ctxt->input->cur - ctxt->input->base;
5913 res = xmlParserInputBufferPush(ctxt->input->buf, size, chunk);
5915 ctxt->errNo = XML_PARSER_EOF;
5916 ctxt->disableSAX = 1;
5917 return (XML_PARSER_EOF);
5919 ctxt->input->base = ctxt->input->buf->buffer->content + base;
5920 ctxt->input->cur = ctxt->input->base + cur;
5922 &ctxt->input->buf->buffer->content[ctxt->input->buf->buffer->use];
5924 xmlGenericError(xmlGenericErrorContext, "HPP: pushed %d\n", size);
5928 if ((terminate) || (ctxt->input->buf->buffer->use > 80))
5929 htmlParseTryOrFinish(ctxt, terminate);
5931 } else if (ctxt->instate != XML_PARSER_EOF) {
5932 if ((ctxt->input != NULL) && ctxt->input->buf != NULL) {
5933 xmlParserInputBufferPtr in = ctxt->input->buf;
5934 if ((in->encoder != NULL) && (in->buffer != NULL) &&
5935 (in->raw != NULL)) {
5938 nbchars = xmlCharEncInFunc(in->encoder, in->buffer, in->raw);
5940 htmlParseErr(ctxt, XML_ERR_INVALID_ENCODING,
5941 "encoder error\n", NULL, NULL);
5942 return(XML_ERR_INVALID_ENCODING);
5947 htmlParseTryOrFinish(ctxt, terminate);
5949 if ((ctxt->instate != XML_PARSER_EOF) &&
5950 (ctxt->instate != XML_PARSER_EPILOG) &&
5951 (ctxt->instate != XML_PARSER_MISC)) {
5952 ctxt->errNo = XML_ERR_DOCUMENT_END;
5953 ctxt->wellFormed = 0;
5955 if (ctxt->instate != XML_PARSER_EOF) {
5956 if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
5957 ctxt->sax->endDocument(ctxt->userData);
5959 ctxt->instate = XML_PARSER_EOF;
5961 return((xmlParserErrors) ctxt->errNo);
5964 /************************************************************************
5966 * User entry points *
5968 ************************************************************************/
5971 * htmlCreatePushParserCtxt:
5972 * @sax: a SAX handler
5973 * @user_data: The user data returned on SAX callbacks
5974 * @chunk: a pointer to an array of chars
5975 * @size: number of chars in the array
5976 * @filename: an optional file name or URI
5977 * @enc: an optional encoding
5979 * Create a parser context for using the HTML parser in push mode
5980 * The value of @filename is used for fetching external entities
5981 * and error/warning reports.
5983 * Returns the new parser context or NULL
5986 htmlCreatePushParserCtxt(htmlSAXHandlerPtr sax, void *user_data,
5987 const char *chunk, int size, const char *filename,
5988 xmlCharEncoding enc) {
5989 htmlParserCtxtPtr ctxt;
5990 htmlParserInputPtr inputStream;
5991 xmlParserInputBufferPtr buf;
5995 buf = xmlAllocParserInputBuffer(enc);
5996 if (buf == NULL) return(NULL);
5998 ctxt = htmlNewParserCtxt();
6000 xmlFreeParserInputBuffer(buf);
6003 if(enc==XML_CHAR_ENCODING_UTF8 || buf->encoder)
6004 ctxt->charset=XML_CHAR_ENCODING_UTF8;
6006 if (ctxt->sax != (xmlSAXHandlerPtr) &htmlDefaultSAXHandler)
6008 ctxt->sax = (htmlSAXHandlerPtr) xmlMalloc(sizeof(htmlSAXHandler));
6009 if (ctxt->sax == NULL) {
6014 memcpy(ctxt->sax, sax, sizeof(htmlSAXHandler));
6015 if (user_data != NULL)
6016 ctxt->userData = user_data;
6018 if (filename == NULL) {
6019 ctxt->directory = NULL;
6021 ctxt->directory = xmlParserGetDirectory(filename);
6024 inputStream = htmlNewInputStream(ctxt);
6025 if (inputStream == NULL) {
6026 xmlFreeParserCtxt(ctxt);
6031 if (filename == NULL)
6032 inputStream->filename = NULL;
6034 inputStream->filename = (char *)
6035 xmlCanonicPath((const xmlChar *) filename);
6036 inputStream->buf = buf;
6037 inputStream->base = inputStream->buf->buffer->content;
6038 inputStream->cur = inputStream->buf->buffer->content;
6040 &inputStream->buf->buffer->content[inputStream->buf->buffer->use];
6042 inputPush(ctxt, inputStream);
6044 if ((size > 0) && (chunk != NULL) && (ctxt->input != NULL) &&
6045 (ctxt->input->buf != NULL)) {
6046 int base = ctxt->input->base - ctxt->input->buf->buffer->content;
6047 int cur = ctxt->input->cur - ctxt->input->base;
6049 xmlParserInputBufferPush(ctxt->input->buf, size, chunk);
6051 ctxt->input->base = ctxt->input->buf->buffer->content + base;
6052 ctxt->input->cur = ctxt->input->base + cur;
6054 &ctxt->input->buf->buffer->content[ctxt->input->buf->buffer->use];
6056 xmlGenericError(xmlGenericErrorContext, "HPP: pushed %d\n", size);
6059 ctxt->progressive = 1;
6063 #endif /* LIBXML_PUSH_ENABLED */
6067 * @cur: a pointer to an array of xmlChar
6068 * @encoding: a free form C string describing the HTML document encoding, or NULL
6069 * @sax: the SAX handler block
6070 * @userData: if using SAX, this pointer will be provided on callbacks.
6072 * Parse an HTML in-memory document. If sax is not NULL, use the SAX callbacks
6073 * to handle parse events. If sax is NULL, fallback to the default DOM
6074 * behavior and return a tree.
6076 * Returns the resulting document tree unless SAX is NULL or the document is
6081 htmlSAXParseDoc(xmlChar *cur, const char *encoding, htmlSAXHandlerPtr sax, void *userData) {
6083 htmlParserCtxtPtr ctxt;
6087 if (cur == NULL) return(NULL);
6090 ctxt = htmlCreateDocParserCtxt(cur, encoding);
6091 if (ctxt == NULL) return(NULL);
6093 if (ctxt->sax != NULL) xmlFree (ctxt->sax);
6095 ctxt->userData = userData;
6098 htmlParseDocument(ctxt);
6102 ctxt->userData = NULL;
6104 htmlFreeParserCtxt(ctxt);
6111 * @cur: a pointer to an array of xmlChar
6112 * @encoding: a free form C string describing the HTML document encoding, or NULL
6114 * parse an HTML in-memory document and build a tree.
6116 * Returns the resulting document tree
6120 htmlParseDoc(xmlChar *cur, const char *encoding) {
6121 return(htmlSAXParseDoc(cur, encoding, NULL, NULL));
6126 * htmlCreateFileParserCtxt:
6127 * @filename: the filename
6128 * @encoding: a free form C string describing the HTML document encoding, or NULL
6130 * Create a parser context for a file content.
6131 * Automatic support for ZLIB/Compress compressed document is provided
6132 * by default if found at compile-time.
6134 * Returns the new parser context or NULL
6137 htmlCreateFileParserCtxt(const char *filename, const char *encoding)
6139 htmlParserCtxtPtr ctxt;
6140 htmlParserInputPtr inputStream;
6141 char *canonicFilename;
6142 /* htmlCharEncoding enc; */
6143 xmlChar *content, *content_line = (xmlChar *) "charset=";
6145 if (filename == NULL)
6148 ctxt = htmlNewParserCtxt();
6152 canonicFilename = (char *) xmlCanonicPath((const xmlChar *) filename);
6153 if (canonicFilename == NULL) {
6154 #ifdef LIBXML_SAX1_ENABLED
6155 if (xmlDefaultSAXHandler.error != NULL) {
6156 xmlDefaultSAXHandler.error(NULL, "out of memory\n");
6159 xmlFreeParserCtxt(ctxt);
6163 inputStream = xmlLoadExternalEntity(canonicFilename, NULL, ctxt);
6164 xmlFree(canonicFilename);
6165 if (inputStream == NULL) {
6166 xmlFreeParserCtxt(ctxt);
6170 inputPush(ctxt, inputStream);
6174 content = xmlMallocAtomic (xmlStrlen(content_line) + strlen(encoding) + 1);
6176 strcpy ((char *)content, (char *)content_line);
6177 strcat ((char *)content, (char *)encoding);
6178 htmlCheckEncoding (ctxt, content);
6188 * @filename: the filename
6189 * @encoding: a free form C string describing the HTML document encoding, or NULL
6190 * @sax: the SAX handler block
6191 * @userData: if using SAX, this pointer will be provided on callbacks.
6193 * parse an HTML file and build a tree. Automatic support for ZLIB/Compress
6194 * compressed document is provided by default if found at compile-time.
6195 * It use the given SAX function block to handle the parsing callback.
6196 * If sax is NULL, fallback to the default DOM tree building routines.
6198 * Returns the resulting document tree unless SAX is NULL or the document is
6203 htmlSAXParseFile(const char *filename, const char *encoding, htmlSAXHandlerPtr sax,
6206 htmlParserCtxtPtr ctxt;
6207 htmlSAXHandlerPtr oldsax = NULL;
6211 ctxt = htmlCreateFileParserCtxt(filename, encoding);
6212 if (ctxt == NULL) return(NULL);
6216 ctxt->userData = userData;
6219 htmlParseDocument(ctxt);
6224 ctxt->userData = NULL;
6226 htmlFreeParserCtxt(ctxt);
6233 * @filename: the filename
6234 * @encoding: a free form C string describing the HTML document encoding, or NULL
6236 * parse an HTML file and build a tree. Automatic support for ZLIB/Compress
6237 * compressed document is provided by default if found at compile-time.
6239 * Returns the resulting document tree
6243 htmlParseFile(const char *filename, const char *encoding) {
6244 return(htmlSAXParseFile(filename, encoding, NULL, NULL));
6248 * htmlHandleOmittedElem:
6251 * Set and return the previous value for handling HTML omitted tags.
6253 * Returns the last value for 0 for no handling, 1 for auto insertion.
6257 htmlHandleOmittedElem(int val) {
6258 int old = htmlOmittedDefaultValue;
6260 htmlOmittedDefaultValue = val;
6265 * htmlElementAllowedHere:
6266 * @parent: HTML parent element
6267 * @elt: HTML element
6269 * Checks whether an HTML element may be a direct child of a parent element.
6270 * Note - doesn't check for deprecated elements
6272 * Returns 1 if allowed; 0 otherwise.
6275 htmlElementAllowedHere(const htmlElemDesc* parent, const xmlChar* elt) {
6278 if ( ! elt || ! parent || ! parent->subelts )
6281 for ( p = parent->subelts; *p; ++p )
6282 if ( !xmlStrcmp((const xmlChar *)*p, elt) )
6288 * htmlElementStatusHere:
6289 * @parent: HTML parent element
6290 * @elt: HTML element
6292 * Checks whether an HTML element may be a direct child of a parent element.
6293 * and if so whether it is valid or deprecated.
6295 * Returns one of HTML_VALID, HTML_DEPRECATED, HTML_INVALID
6298 htmlElementStatusHere(const htmlElemDesc* parent, const htmlElemDesc* elt) {
6299 if ( ! parent || ! elt )
6300 return HTML_INVALID ;
6301 if ( ! htmlElementAllowedHere(parent, (const xmlChar*) elt->name ) )
6302 return HTML_INVALID ;
6304 return ( elt->dtd == 0 ) ? HTML_VALID : HTML_DEPRECATED ;
6308 * @elt: HTML element
6309 * @attr: HTML attribute
6310 * @legacy: whether to allow deprecated attributes
6312 * Checks whether an attribute is valid for an element
6313 * Has full knowledge of Required and Deprecated attributes
6315 * Returns one of HTML_REQUIRED, HTML_VALID, HTML_DEPRECATED, HTML_INVALID
6318 htmlAttrAllowed(const htmlElemDesc* elt, const xmlChar* attr, int legacy) {
6321 if ( !elt || ! attr )
6322 return HTML_INVALID ;
6324 if ( elt->attrs_req )
6325 for ( p = elt->attrs_req; *p; ++p)
6326 if ( !xmlStrcmp((const xmlChar*)*p, attr) )
6327 return HTML_REQUIRED ;
6329 if ( elt->attrs_opt )
6330 for ( p = elt->attrs_opt; *p; ++p)
6331 if ( !xmlStrcmp((const xmlChar*)*p, attr) )
6334 if ( legacy && elt->attrs_depr )
6335 for ( p = elt->attrs_depr; *p; ++p)
6336 if ( !xmlStrcmp((const xmlChar*)*p, attr) )
6337 return HTML_DEPRECATED ;
6339 return HTML_INVALID ;
6343 * @node: an htmlNodePtr in a tree
6344 * @legacy: whether to allow deprecated elements (YES is faster here
6345 * for Element nodes)
6347 * Checks whether the tree node is valid. Experimental (the author
6348 * only uses the HTML enhancements in a SAX parser)
6350 * Return: for Element nodes, a return from htmlElementAllowedHere (if
6351 * legacy allowed) or htmlElementStatusHere (otherwise).
6352 * for Attribute nodes, a return from htmlAttrAllowed
6353 * for other nodes, HTML_NA (no checks performed)
6356 htmlNodeStatus(const htmlNodePtr node, int legacy) {
6358 return HTML_INVALID ;
6360 switch ( node->type ) {
6361 case XML_ELEMENT_NODE:
6363 ? ( htmlElementAllowedHere (
6364 htmlTagLookup(node->parent->name) , node->name
6365 ) ? HTML_VALID : HTML_INVALID )
6366 : htmlElementStatusHere(
6367 htmlTagLookup(node->parent->name) ,
6368 htmlTagLookup(node->name) )
6370 case XML_ATTRIBUTE_NODE:
6371 return htmlAttrAllowed(
6372 htmlTagLookup(node->parent->name) , node->name, legacy) ;
6373 default: return HTML_NA ;
6376 /************************************************************************
6378 * New set (2.6.0) of simpler and more flexible APIs *
6380 ************************************************************************/
6385 * Free a string if it is not owned by the "dict" dictionnary in the
6388 #define DICT_FREE(str) \
6389 if ((str) && ((!dict) || \
6390 (xmlDictOwns(dict, (const xmlChar *)(str)) == 0))) \
6391 xmlFree((char *)(str));
6395 * @ctxt: an HTML parser context
6397 * Reset a parser context
6400 htmlCtxtReset(htmlParserCtxtPtr ctxt)
6402 xmlParserInputPtr input;
6411 while ((input = inputPop(ctxt)) != NULL) { /* Non consuming */
6412 xmlFreeInputStream(input);
6418 if (ctxt->spaceTab != NULL) {
6419 ctxt->spaceTab[0] = -1;
6420 ctxt->space = &ctxt->spaceTab[0];
6432 DICT_FREE(ctxt->version);
6433 ctxt->version = NULL;
6434 DICT_FREE(ctxt->encoding);
6435 ctxt->encoding = NULL;
6436 DICT_FREE(ctxt->directory);
6437 ctxt->directory = NULL;
6438 DICT_FREE(ctxt->extSubURI);
6439 ctxt->extSubURI = NULL;
6440 DICT_FREE(ctxt->extSubSystem);
6441 ctxt->extSubSystem = NULL;
6442 if (ctxt->myDoc != NULL)
6443 xmlFreeDoc(ctxt->myDoc);
6446 ctxt->standalone = -1;
6447 ctxt->hasExternalSubset = 0;
6448 ctxt->hasPErefs = 0;
6451 ctxt->instate = XML_PARSER_START;
6454 ctxt->wellFormed = 1;
6455 ctxt->nsWellFormed = 1;
6456 ctxt->disableSAX = 0;
6458 ctxt->vctxt.userData = ctxt;
6459 ctxt->vctxt.error = xmlParserValidityError;
6460 ctxt->vctxt.warning = xmlParserValidityWarning;
6461 ctxt->record_info = 0;
6463 ctxt->checkIndex = 0;
6465 ctxt->errNo = XML_ERR_OK;
6467 ctxt->charset = XML_CHAR_ENCODING_NONE;
6468 ctxt->catalogs = NULL;
6469 xmlInitNodeInfoSeq(&ctxt->node_seq);
6471 if (ctxt->attsDefault != NULL) {
6472 xmlHashFree(ctxt->attsDefault, (xmlHashDeallocator) xmlFree);
6473 ctxt->attsDefault = NULL;
6475 if (ctxt->attsSpecial != NULL) {
6476 xmlHashFree(ctxt->attsSpecial, NULL);
6477 ctxt->attsSpecial = NULL;
6482 * htmlCtxtUseOptions:
6483 * @ctxt: an HTML parser context
6484 * @options: a combination of htmlParserOption(s)
6486 * Applies the options to the parser context
6488 * Returns 0 in case of success, the set of unknown or unimplemented options
6492 htmlCtxtUseOptions(htmlParserCtxtPtr ctxt, int options)
6497 if (options & HTML_PARSE_NOWARNING) {
6498 ctxt->sax->warning = NULL;
6499 ctxt->vctxt.warning = NULL;
6500 options -= XML_PARSE_NOWARNING;
6501 ctxt->options |= XML_PARSE_NOWARNING;
6503 if (options & HTML_PARSE_NOERROR) {
6504 ctxt->sax->error = NULL;
6505 ctxt->vctxt.error = NULL;
6506 ctxt->sax->fatalError = NULL;
6507 options -= XML_PARSE_NOERROR;
6508 ctxt->options |= XML_PARSE_NOERROR;
6510 if (options & HTML_PARSE_PEDANTIC) {
6512 options -= XML_PARSE_PEDANTIC;
6513 ctxt->options |= XML_PARSE_PEDANTIC;
6516 if (options & XML_PARSE_NOBLANKS) {
6517 ctxt->keepBlanks = 0;
6518 ctxt->sax->ignorableWhitespace = xmlSAX2IgnorableWhitespace;
6519 options -= XML_PARSE_NOBLANKS;
6520 ctxt->options |= XML_PARSE_NOBLANKS;
6522 ctxt->keepBlanks = 1;
6523 if (options & HTML_PARSE_RECOVER) {
6525 options -= HTML_PARSE_RECOVER;
6528 if (options & HTML_PARSE_COMPACT) {
6529 ctxt->options |= HTML_PARSE_COMPACT;
6530 options -= HTML_PARSE_COMPACT;
6532 if (options & XML_PARSE_HUGE) {
6533 ctxt->options |= XML_PARSE_HUGE;
6534 options -= XML_PARSE_HUGE;
6536 if (options & HTML_PARSE_NODEFDTD) {
6537 ctxt->options |= HTML_PARSE_NODEFDTD;
6538 options -= HTML_PARSE_NODEFDTD;
6540 ctxt->dictNames = 0;
6546 * @ctxt: an HTML parser context
6547 * @URL: the base URL to use for the document
6548 * @encoding: the document encoding, or NULL
6549 * @options: a combination of htmlParserOption(s)
6550 * @reuse: keep the context for reuse
6552 * Common front-end for the htmlRead functions
6554 * Returns the resulting document tree or NULL
6557 htmlDoRead(htmlParserCtxtPtr ctxt, const char *URL, const char *encoding,
6558 int options, int reuse)
6562 htmlCtxtUseOptions(ctxt, options);
6564 if (encoding != NULL) {
6565 xmlCharEncodingHandlerPtr hdlr;
6567 hdlr = xmlFindCharEncodingHandler(encoding);
6569 xmlSwitchToEncoding(ctxt, hdlr);
6570 if (ctxt->input->encoding != NULL)
6571 xmlFree((xmlChar *) ctxt->input->encoding);
6572 ctxt->input->encoding = xmlStrdup((xmlChar *)encoding);
6575 if ((URL != NULL) && (ctxt->input != NULL) &&
6576 (ctxt->input->filename == NULL))
6577 ctxt->input->filename = (char *) xmlStrdup((const xmlChar *) URL);
6578 htmlParseDocument(ctxt);
6582 if ((ctxt->dictNames) &&
6584 (ret->dict == ctxt->dict))
6586 xmlFreeParserCtxt(ctxt);
6593 * @cur: a pointer to a zero terminated string
6594 * @URL: the base URL to use for the document
6595 * @encoding: the document encoding, or NULL
6596 * @options: a combination of htmlParserOption(s)
6598 * parse an XML in-memory document and build a tree.
6600 * Returns the resulting document tree
6603 htmlReadDoc(const xmlChar * cur, const char *URL, const char *encoding, int options)
6605 htmlParserCtxtPtr ctxt;
6611 ctxt = htmlCreateDocParserCtxt(cur, NULL);
6614 return (htmlDoRead(ctxt, URL, encoding, options, 0));
6619 * @filename: a file or URL
6620 * @encoding: the document encoding, or NULL
6621 * @options: a combination of htmlParserOption(s)
6623 * parse an XML file from the filesystem or the network.
6625 * Returns the resulting document tree
6628 htmlReadFile(const char *filename, const char *encoding, int options)
6630 htmlParserCtxtPtr ctxt;
6633 ctxt = htmlCreateFileParserCtxt(filename, encoding);
6636 return (htmlDoRead(ctxt, NULL, NULL, options, 0));
6641 * @buffer: a pointer to a char array
6642 * @size: the size of the array
6643 * @URL: the base URL to use for the document
6644 * @encoding: the document encoding, or NULL
6645 * @options: a combination of htmlParserOption(s)
6647 * parse an XML in-memory document and build a tree.
6649 * Returns the resulting document tree
6652 htmlReadMemory(const char *buffer, int size, const char *URL, const char *encoding, int options)
6654 htmlParserCtxtPtr ctxt;
6657 ctxt = xmlCreateMemoryParserCtxt(buffer, size);
6660 htmlDefaultSAXHandlerInit();
6661 if (ctxt->sax != NULL)
6662 memcpy(ctxt->sax, &htmlDefaultSAXHandler, sizeof(xmlSAXHandlerV1));
6663 return (htmlDoRead(ctxt, URL, encoding, options, 0));
6668 * @fd: an open file descriptor
6669 * @URL: the base URL to use for the document
6670 * @encoding: the document encoding, or NULL
6671 * @options: a combination of htmlParserOption(s)
6673 * parse an XML from a file descriptor and build a tree.
6675 * Returns the resulting document tree
6678 htmlReadFd(int fd, const char *URL, const char *encoding, int options)
6680 htmlParserCtxtPtr ctxt;
6681 xmlParserInputBufferPtr input;
6682 xmlParserInputPtr stream;
6688 input = xmlParserInputBufferCreateFd(fd, XML_CHAR_ENCODING_NONE);
6691 ctxt = xmlNewParserCtxt();
6693 xmlFreeParserInputBuffer(input);
6696 stream = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE);
6697 if (stream == NULL) {
6698 xmlFreeParserInputBuffer(input);
6699 xmlFreeParserCtxt(ctxt);
6702 inputPush(ctxt, stream);
6703 return (htmlDoRead(ctxt, URL, encoding, options, 0));
6708 * @ioread: an I/O read function
6709 * @ioclose: an I/O close function
6710 * @ioctx: an I/O handler
6711 * @URL: the base URL to use for the document
6712 * @encoding: the document encoding, or NULL
6713 * @options: a combination of htmlParserOption(s)
6715 * parse an HTML document from I/O functions and source and build a tree.
6717 * Returns the resulting document tree
6720 htmlReadIO(xmlInputReadCallback ioread, xmlInputCloseCallback ioclose,
6721 void *ioctx, const char *URL, const char *encoding, int options)
6723 htmlParserCtxtPtr ctxt;
6724 xmlParserInputBufferPtr input;
6725 xmlParserInputPtr stream;
6731 input = xmlParserInputBufferCreateIO(ioread, ioclose, ioctx,
6732 XML_CHAR_ENCODING_NONE);
6735 ctxt = htmlNewParserCtxt();
6737 xmlFreeParserInputBuffer(input);
6740 stream = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE);
6741 if (stream == NULL) {
6742 xmlFreeParserInputBuffer(input);
6743 xmlFreeParserCtxt(ctxt);
6746 inputPush(ctxt, stream);
6747 return (htmlDoRead(ctxt, URL, encoding, options, 0));
6752 * @ctxt: an HTML parser context
6753 * @cur: a pointer to a zero terminated string
6754 * @URL: the base URL to use for the document
6755 * @encoding: the document encoding, or NULL
6756 * @options: a combination of htmlParserOption(s)
6758 * parse an XML in-memory document and build a tree.
6759 * This reuses the existing @ctxt parser context
6761 * Returns the resulting document tree
6764 htmlCtxtReadDoc(htmlParserCtxtPtr ctxt, const xmlChar * cur,
6765 const char *URL, const char *encoding, int options)
6767 xmlParserInputPtr stream;
6774 htmlCtxtReset(ctxt);
6776 stream = xmlNewStringInputStream(ctxt, cur);
6777 if (stream == NULL) {
6780 inputPush(ctxt, stream);
6781 return (htmlDoRead(ctxt, URL, encoding, options, 1));
6786 * @ctxt: an HTML parser context
6787 * @filename: a file or URL
6788 * @encoding: the document encoding, or NULL
6789 * @options: a combination of htmlParserOption(s)
6791 * parse an XML file from the filesystem or the network.
6792 * This reuses the existing @ctxt parser context
6794 * Returns the resulting document tree
6797 htmlCtxtReadFile(htmlParserCtxtPtr ctxt, const char *filename,
6798 const char *encoding, int options)
6800 xmlParserInputPtr stream;
6802 if (filename == NULL)
6807 htmlCtxtReset(ctxt);
6809 stream = xmlLoadExternalEntity(filename, NULL, ctxt);
6810 if (stream == NULL) {
6813 inputPush(ctxt, stream);
6814 return (htmlDoRead(ctxt, NULL, encoding, options, 1));
6818 * htmlCtxtReadMemory:
6819 * @ctxt: an HTML parser context
6820 * @buffer: a pointer to a char array
6821 * @size: the size of the array
6822 * @URL: the base URL to use for the document
6823 * @encoding: the document encoding, or NULL
6824 * @options: a combination of htmlParserOption(s)
6826 * parse an XML in-memory document and build a tree.
6827 * This reuses the existing @ctxt parser context
6829 * Returns the resulting document tree
6832 htmlCtxtReadMemory(htmlParserCtxtPtr ctxt, const char *buffer, int size,
6833 const char *URL, const char *encoding, int options)
6835 xmlParserInputBufferPtr input;
6836 xmlParserInputPtr stream;
6843 htmlCtxtReset(ctxt);
6845 input = xmlParserInputBufferCreateMem(buffer, size, XML_CHAR_ENCODING_NONE);
6846 if (input == NULL) {
6850 stream = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE);
6851 if (stream == NULL) {
6852 xmlFreeParserInputBuffer(input);
6856 inputPush(ctxt, stream);
6857 return (htmlDoRead(ctxt, URL, encoding, options, 1));
6862 * @ctxt: an HTML parser context
6863 * @fd: an open file descriptor
6864 * @URL: the base URL to use for the document
6865 * @encoding: the document encoding, or NULL
6866 * @options: a combination of htmlParserOption(s)
6868 * parse an XML from a file descriptor and build a tree.
6869 * This reuses the existing @ctxt parser context
6871 * Returns the resulting document tree
6874 htmlCtxtReadFd(htmlParserCtxtPtr ctxt, int fd,
6875 const char *URL, const char *encoding, int options)
6877 xmlParserInputBufferPtr input;
6878 xmlParserInputPtr stream;
6885 htmlCtxtReset(ctxt);
6888 input = xmlParserInputBufferCreateFd(fd, XML_CHAR_ENCODING_NONE);
6891 stream = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE);
6892 if (stream == NULL) {
6893 xmlFreeParserInputBuffer(input);
6896 inputPush(ctxt, stream);
6897 return (htmlDoRead(ctxt, URL, encoding, options, 1));
6902 * @ctxt: an HTML parser context
6903 * @ioread: an I/O read function
6904 * @ioclose: an I/O close function
6905 * @ioctx: an I/O handler
6906 * @URL: the base URL to use for the document
6907 * @encoding: the document encoding, or NULL
6908 * @options: a combination of htmlParserOption(s)
6910 * parse an HTML document from I/O functions and source and build a tree.
6911 * This reuses the existing @ctxt parser context
6913 * Returns the resulting document tree
6916 htmlCtxtReadIO(htmlParserCtxtPtr ctxt, xmlInputReadCallback ioread,
6917 xmlInputCloseCallback ioclose, void *ioctx,
6919 const char *encoding, int options)
6921 xmlParserInputBufferPtr input;
6922 xmlParserInputPtr stream;
6929 htmlCtxtReset(ctxt);
6931 input = xmlParserInputBufferCreateIO(ioread, ioclose, ioctx,
6932 XML_CHAR_ENCODING_NONE);
6935 stream = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE);
6936 if (stream == NULL) {
6937 xmlFreeParserInputBuffer(input);
6940 inputPush(ctxt, stream);
6941 return (htmlDoRead(ctxt, URL, encoding, options, 1));
6944 #define bottom_HTMLparser
6945 #include "elfgcchack.h"
6946 #endif /* LIBXML_HTML_ENABLED */