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>
50 #define HTML_MAX_NAMELEN 1000
51 #define HTML_PARSER_BIG_BUFFER_SIZE 1000
52 #define HTML_PARSER_BUFFER_SIZE 100
55 /* #define DEBUG_PUSH */
57 static int htmlOmittedDefaultValue = 1;
59 xmlChar * htmlDecodeEntities(htmlParserCtxtPtr ctxt, int len,
60 xmlChar end, xmlChar end2, xmlChar end3);
61 static void htmlParseComment(htmlParserCtxtPtr ctxt);
63 /************************************************************************
65 * Some factorized error routines *
67 ************************************************************************/
71 * @ctxt: an HTML parser context
72 * @extra: extra informations
74 * Handle a redefinition of attribute error
77 htmlErrMemory(xmlParserCtxtPtr ctxt, const char *extra)
79 if ((ctxt != NULL) && (ctxt->disableSAX != 0) &&
80 (ctxt->instate == XML_PARSER_EOF))
83 ctxt->errNo = XML_ERR_NO_MEMORY;
84 ctxt->instate = XML_PARSER_EOF;
88 __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER,
89 XML_ERR_NO_MEMORY, XML_ERR_FATAL, NULL, 0, extra,
91 "Memory allocation failed : %s\n", extra);
93 __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER,
94 XML_ERR_NO_MEMORY, XML_ERR_FATAL, NULL, 0, NULL,
95 NULL, NULL, 0, 0, "Memory allocation failed\n");
100 * @ctxt: an HTML parser context
101 * @error: the error number
102 * @msg: the error message
103 * @str1: string infor
104 * @str2: string infor
106 * Handle a fatal parser error, i.e. violating Well-Formedness constraints
109 htmlParseErr(xmlParserCtxtPtr ctxt, xmlParserErrors error,
110 const char *msg, const xmlChar *str1, const xmlChar *str2)
112 if ((ctxt != NULL) && (ctxt->disableSAX != 0) &&
113 (ctxt->instate == XML_PARSER_EOF))
117 __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_HTML, error,
118 XML_ERR_ERROR, NULL, 0,
119 (const char *) str1, (const char *) str2,
123 ctxt->wellFormed = 0;
128 * @ctxt: an HTML parser context
129 * @error: the error number
130 * @msg: the error message
133 * Handle a fatal parser error, i.e. violating Well-Formedness constraints
136 htmlParseErrInt(xmlParserCtxtPtr ctxt, xmlParserErrors error,
137 const char *msg, int val)
139 if ((ctxt != NULL) && (ctxt->disableSAX != 0) &&
140 (ctxt->instate == XML_PARSER_EOF))
144 __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_HTML, error,
145 XML_ERR_ERROR, NULL, 0, NULL, NULL,
146 NULL, val, 0, msg, val);
148 ctxt->wellFormed = 0;
151 /************************************************************************
153 * Parser stacks related functions and macros *
155 ************************************************************************/
159 * @ctxt: an HTML parser context
160 * @value: the element name
162 * Pushes a new element name on top of the name stack
164 * Returns 0 in case of error, the index in the stack otherwise
167 htmlnamePush(htmlParserCtxtPtr ctxt, const xmlChar * value)
169 if ((ctxt->html < 3) && (xmlStrEqual(value, BAD_CAST "head")))
171 if ((ctxt->html < 10) && (xmlStrEqual(value, BAD_CAST "body")))
173 if (ctxt->nameNr >= ctxt->nameMax) {
175 ctxt->nameTab = (const xmlChar * *)
176 xmlRealloc((xmlChar * *)ctxt->nameTab,
178 sizeof(ctxt->nameTab[0]));
179 if (ctxt->nameTab == NULL) {
180 htmlErrMemory(ctxt, NULL);
184 ctxt->nameTab[ctxt->nameNr] = value;
186 return (ctxt->nameNr++);
190 * @ctxt: an HTML parser context
192 * Pops the top element name from the name stack
194 * Returns the name just removed
196 static const xmlChar *
197 htmlnamePop(htmlParserCtxtPtr ctxt)
201 if (ctxt->nameNr <= 0)
204 if (ctxt->nameNr < 0)
206 if (ctxt->nameNr > 0)
207 ctxt->name = ctxt->nameTab[ctxt->nameNr - 1];
210 ret = ctxt->nameTab[ctxt->nameNr];
211 ctxt->nameTab[ctxt->nameNr] = NULL;
217 * @ctxt: an HTML parser context
218 * @value: the node info
220 * Pushes a new element name on top of the node info stack
222 * Returns 0 in case of error, the index in the stack otherwise
225 htmlNodeInfoPush(htmlParserCtxtPtr ctxt, htmlParserNodeInfo *value)
227 if (ctxt->nodeInfoNr >= ctxt->nodeInfoMax) {
228 if (ctxt->nodeInfoMax == 0)
229 ctxt->nodeInfoMax = 5;
230 ctxt->nodeInfoMax *= 2;
231 ctxt->nodeInfoTab = (htmlParserNodeInfo *)
232 xmlRealloc((htmlParserNodeInfo *)ctxt->nodeInfoTab,
234 sizeof(ctxt->nodeInfoTab[0]));
235 if (ctxt->nodeInfoTab == NULL) {
236 htmlErrMemory(ctxt, NULL);
240 ctxt->nodeInfoTab[ctxt->nodeInfoNr] = *value;
241 ctxt->nodeInfo = &ctxt->nodeInfoTab[ctxt->nodeInfoNr];
242 return (ctxt->nodeInfoNr++);
247 * @ctxt: an HTML parser context
249 * Pops the top element name from the node info stack
251 * Returns 0 in case of error, the pointer to NodeInfo otherwise
253 static htmlParserNodeInfo *
254 htmlNodeInfoPop(htmlParserCtxtPtr ctxt)
256 if (ctxt->nodeInfoNr <= 0)
259 if (ctxt->nodeInfoNr < 0)
261 if (ctxt->nodeInfoNr > 0)
262 ctxt->nodeInfo = &ctxt->nodeInfoTab[ctxt->nodeInfoNr - 1];
264 ctxt->nodeInfo = NULL;
265 return &ctxt->nodeInfoTab[ctxt->nodeInfoNr];
269 * Macros for accessing the content. Those should be used only by the parser,
272 * Dirty macros, i.e. one need to make assumption on the context to use them
274 * CUR_PTR return the current pointer to the xmlChar to be parsed.
275 * CUR returns the current xmlChar value, i.e. a 8 bit value if compiled
276 * in ISO-Latin or UTF-8, and the current 16 bit value if compiled
277 * in UNICODE mode. This should be used internally by the parser
278 * only to compare to ASCII values otherwise it would break when
279 * running with UTF-8 encoding.
280 * NXT(n) returns the n'th next xmlChar. Same as CUR is should be used only
281 * to compare on ASCII based substring.
282 * UPP(n) returns the n'th next xmlChar converted to uppercase. Same as CUR
283 * it should be used only to compare on ASCII based substring.
284 * SKIP(n) Skip n xmlChar, and must also be used only to skip ASCII defined
285 * strings without newlines within the parser.
287 * Clean macros, not dependent of an ASCII context, expect UTF-8 encoding
289 * CURRENT Returns the current char value, with the full decoding of
290 * UTF-8 if we are using this mode. It returns an int.
291 * NEXT Skip to the next character, this does the proper decoding
292 * in UTF-8 mode. It also pop-up unfinished entities on the fly.
293 * NEXTL(l) Skip the current unicode character of l xmlChars long.
294 * COPY(to) copy one char to *to, increment CUR_PTR and to accordingly
297 #define UPPER (toupper(*ctxt->input->cur))
299 #define SKIP(val) ctxt->nbChars += (val),ctxt->input->cur += (val),ctxt->input->col+=(val)
301 #define NXT(val) ctxt->input->cur[(val)]
303 #define UPP(val) (toupper(ctxt->input->cur[(val)]))
305 #define CUR_PTR ctxt->input->cur
307 #define SHRINK if ((ctxt->input->cur - ctxt->input->base > 2 * INPUT_CHUNK) && \
308 (ctxt->input->end - ctxt->input->cur < 2 * INPUT_CHUNK)) \
309 xmlParserInputShrink(ctxt->input)
311 #define GROW if ((ctxt->progressive == 0) && \
312 (ctxt->input->end - ctxt->input->cur < INPUT_CHUNK)) \
313 xmlParserInputGrow(ctxt->input, INPUT_CHUNK)
315 #define CURRENT ((int) (*ctxt->input->cur))
317 #define SKIP_BLANKS htmlSkipBlankChars(ctxt)
319 /* Inported from XML */
321 /* #define CUR (ctxt->token ? ctxt->token : (int) (*ctxt->input->cur)) */
322 #define CUR ((int) (*ctxt->input->cur))
323 #define NEXT xmlNextChar(ctxt)
325 #define RAW (ctxt->token ? -1 : (*ctxt->input->cur))
328 #define NEXTL(l) do { \
329 if (*(ctxt->input->cur) == '\n') { \
330 ctxt->input->line++; ctxt->input->col = 1; \
331 } else ctxt->input->col++; \
332 ctxt->token = 0; ctxt->input->cur += l; ctxt->nbChars++; \
337 if (*ctxt->input->cur == '%') xmlParserHandlePEReference(ctxt); \
338 if (*ctxt->input->cur == '&') xmlParserHandleReference(ctxt);
341 #define CUR_CHAR(l) htmlCurrentChar(ctxt, &l)
342 #define CUR_SCHAR(s, l) xmlStringCurrentChar(ctxt, s, &l)
344 #define COPY_BUF(l,b,i,v) \
345 if (l == 1) b[i++] = (xmlChar) v; \
346 else i += xmlCopyChar(l,&b[i],v)
350 * @the HTML parser context
352 * Ty to find and encoding in the current data available in the input
353 * buffer this is needed to try to switch to the proper encoding when
354 * one face a character error.
355 * That's an heuristic, since it's operating outside of parsing it could
356 * try to use a meta which had been commented out, that's the reason it
357 * should only be used in case of error, not as a default.
359 * Returns an encoding string or NULL if not found, the string need to
363 htmlFindEncoding(xmlParserCtxtPtr ctxt) {
364 const xmlChar *start, *cur, *end;
366 if ((ctxt == NULL) || (ctxt->input == NULL) ||
367 (ctxt->input->encoding != NULL) || (ctxt->input->buf == NULL) ||
368 (ctxt->input->buf->encoder != NULL))
370 if ((ctxt->input->cur == NULL) || (ctxt->input->end == NULL))
373 start = ctxt->input->cur;
374 end = ctxt->input->end;
375 /* we also expect the input buffer to be zero terminated */
379 cur = xmlStrcasestr(start, BAD_CAST "HTTP-EQUIV");
382 cur = xmlStrcasestr(cur, BAD_CAST "CONTENT");
385 cur = xmlStrcasestr(cur, BAD_CAST "CHARSET=");
390 while (((*cur >= 'A') && (*cur <= 'Z')) ||
391 ((*cur >= 'a') && (*cur <= 'z')) ||
392 ((*cur >= '0') && (*cur <= '9')) ||
393 (*cur == '-') || (*cur == '_') || (*cur == ':') || (*cur == '/'))
397 return(xmlStrndup(start, cur - start));
402 * @ctxt: the HTML parser context
403 * @len: pointer to the length of the char read
405 * The current char value, if using UTF-8 this may actually span multiple
406 * bytes in the input buffer. Implement the end of line normalization:
407 * 2.11 End-of-Line Handling
408 * If the encoding is unspecified, in the case we find an ISO-Latin-1
409 * char, then the encoding converter is plugged in automatically.
411 * Returns the current char value and its length
415 htmlCurrentChar(xmlParserCtxtPtr ctxt, int *len) {
416 if (ctxt->instate == XML_PARSER_EOF)
419 if (ctxt->token != 0) {
423 if (ctxt->charset == XML_CHAR_ENCODING_UTF8) {
425 * We are supposed to handle UTF8, check it's valid
426 * From rfc2044: encoding of the Unicode values on UTF-8:
428 * UCS-4 range (hex.) UTF-8 octet sequence (binary)
429 * 0000 0000-0000 007F 0xxxxxxx
430 * 0000 0080-0000 07FF 110xxxxx 10xxxxxx
431 * 0000 0800-0000 FFFF 1110xxxx 10xxxxxx 10xxxxxx
433 * Check for the 0x110000 limit too
435 const unsigned char *cur = ctxt->input->cur;
442 xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
443 cur = ctxt->input->cur;
445 if ((cur[1] & 0xc0) != 0x80)
447 if ((c & 0xe0) == 0xe0) {
450 xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
451 cur = ctxt->input->cur;
453 if ((cur[2] & 0xc0) != 0x80)
455 if ((c & 0xf0) == 0xf0) {
457 xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
458 cur = ctxt->input->cur;
460 if (((c & 0xf8) != 0xf0) ||
461 ((cur[3] & 0xc0) != 0x80))
465 val = (cur[0] & 0x7) << 18;
466 val |= (cur[1] & 0x3f) << 12;
467 val |= (cur[2] & 0x3f) << 6;
468 val |= cur[3] & 0x3f;
472 val = (cur[0] & 0xf) << 12;
473 val |= (cur[1] & 0x3f) << 6;
474 val |= cur[2] & 0x3f;
479 val = (cur[0] & 0x1f) << 6;
480 val |= cur[1] & 0x3f;
483 htmlParseErrInt(ctxt, XML_ERR_INVALID_CHAR,
484 "Char 0x%X out of allowed range\n", val);
488 if ((*ctxt->input->cur == 0) &&
489 (ctxt->input->cur < ctxt->input->end)) {
490 htmlParseErrInt(ctxt, XML_ERR_INVALID_CHAR,
491 "Char 0x%X out of allowed range\n", 0);
497 return((int) *ctxt->input->cur);
501 * Assume it's a fixed length encoding (1) with
502 * a compatible encoding for the ASCII set, since
503 * XML constructs only use < 128 chars
506 if ((int) *ctxt->input->cur < 0x80)
507 return((int) *ctxt->input->cur);
510 * Humm this is bad, do an automatic flow conversion
514 xmlCharEncodingHandlerPtr handler;
516 guess = htmlFindEncoding(ctxt);
518 xmlSwitchEncoding(ctxt, XML_CHAR_ENCODING_8859_1);
520 if (ctxt->input->encoding != NULL)
521 xmlFree((xmlChar *) ctxt->input->encoding);
522 ctxt->input->encoding = guess;
523 handler = xmlFindCharEncodingHandler((const char *) guess);
524 if (handler != NULL) {
525 xmlSwitchToEncoding(ctxt, handler);
527 htmlParseErr(ctxt, XML_ERR_INVALID_ENCODING,
528 "Unsupported encoding %s", guess, NULL);
531 ctxt->charset = XML_CHAR_ENCODING_UTF8;
534 return(xmlCurrentChar(ctxt, len));
538 * If we detect an UTF8 error that probably mean that the
539 * input encoding didn't get properly advertized in the
540 * declaration header. Report the error and switch the encoding
541 * to ISO-Latin-1 (if you don't like this policy, just declare the
547 if (ctxt->input->end - ctxt->input->cur >= 4) {
548 snprintf(buffer, 149, "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n",
549 ctxt->input->cur[0], ctxt->input->cur[1],
550 ctxt->input->cur[2], ctxt->input->cur[3]);
552 snprintf(buffer, 149, "Bytes: 0x%02X\n", ctxt->input->cur[0]);
554 htmlParseErr(ctxt, XML_ERR_INVALID_ENCODING,
555 "Input is not proper UTF-8, indicate encoding !\n",
556 BAD_CAST buffer, NULL);
559 ctxt->charset = XML_CHAR_ENCODING_8859_1;
561 return((int) *ctxt->input->cur);
565 * htmlSkipBlankChars:
566 * @ctxt: the HTML parser context
568 * skip all blanks character found at that point in the input streams.
570 * Returns the number of space chars skipped
574 htmlSkipBlankChars(xmlParserCtxtPtr ctxt) {
577 while (IS_BLANK_CH(*(ctxt->input->cur))) {
578 if ((*ctxt->input->cur == 0) &&
579 (xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0)) {
582 if (*(ctxt->input->cur) == '\n') {
583 ctxt->input->line++; ctxt->input->col = 1;
584 } else ctxt->input->col++;
587 if (*ctxt->input->cur == 0)
588 xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
597 /************************************************************************
599 * The list of HTML elements and their properties *
601 ************************************************************************/
604 * Start Tag: 1 means the start tag can be ommited
605 * End Tag: 1 means the end tag can be ommited
606 * 2 means it's forbidden (empty elements)
607 * 3 means the tag is stylistic and should be closed easily
608 * Depr: this element is deprecated
609 * DTD: 1 means that this element is valid only in the Loose DTD
610 * 2 means that this element is valid only in the Frameset DTD
612 * Name,Start Tag,End Tag,Save End,Empty,Deprecated,DTD,inline,Description
613 , subElements , impliedsubelt , Attributes, userdata
616 /* Definitions and a couple of vars for HTML Elements */
618 #define FONTSTYLE "tt", "i", "b", "u", "s", "strike", "big", "small"
619 #define NB_FONTSTYLE 8
620 #define PHRASE "em", "strong", "dfn", "code", "samp", "kbd", "var", "cite", "abbr", "acronym"
622 #define SPECIAL "a", "img", "applet", "embed", "object", "font", "basefont", "br", "script", "map", "q", "sub", "sup", "span", "bdo", "iframe"
623 #define NB_SPECIAL 16
624 #define INLINE FONTSTYLE, PHRASE, SPECIAL, FORMCTRL
625 #define NB_INLINE NB_PCDATA + NB_FONTSTYLE + NB_PHRASE + NB_SPECIAL + NB_FORMCTRL
626 #define BLOCK HEADING, LIST, "pre", "p", "dl", "div", "center", "noscript", "noframes", "blockquote", "form", "isindex", "hr", "table", "fieldset", "address"
627 #define NB_BLOCK NB_HEADING + NB_LIST + 14
628 #define FORMCTRL "input", "select", "textarea", "label", "button"
629 #define NB_FORMCTRL 5
632 #define HEADING "h1", "h2", "h3", "h4", "h5", "h6"
634 #define LIST "ul", "ol", "dir", "menu"
637 #define NB_MODIFIER 0
638 #define FLOW BLOCK,INLINE
639 #define NB_FLOW NB_BLOCK + NB_INLINE
643 static const char* const html_flow[] = { FLOW, NULL } ;
644 static const char* const html_inline[] = { INLINE, NULL } ;
646 /* placeholders: elts with content but no subelements */
647 static const char* const html_pcdata[] = { NULL } ;
648 #define html_cdata html_pcdata
651 /* ... and for HTML Attributes */
653 #define COREATTRS "id", "class", "style", "title"
654 #define NB_COREATTRS 4
655 #define I18N "lang", "dir"
657 #define EVENTS "onclick", "ondblclick", "onmousedown", "onmouseup", "onmouseover", "onmouseout", "onkeypress", "onkeydown", "onkeyup"
659 #define ATTRS COREATTRS,I18N,EVENTS
660 #define NB_ATTRS NB_NB_COREATTRS + NB_I18N + NB_EVENTS
661 #define CELLHALIGN "align", "char", "charoff"
662 #define NB_CELLHALIGN 3
663 #define CELLVALIGN "valign"
664 #define NB_CELLVALIGN 1
666 static const char* const html_attrs[] = { ATTRS, NULL } ;
667 static const char* const core_i18n_attrs[] = { COREATTRS, I18N, NULL } ;
668 static const char* const core_attrs[] = { COREATTRS, NULL } ;
669 static const char* const i18n_attrs[] = { I18N, NULL } ;
672 /* Other declarations that should go inline ... */
673 static const char* const a_attrs[] = { ATTRS, "charset", "type", "name",
674 "href", "hreflang", "rel", "rev", "accesskey", "shape", "coords",
675 "tabindex", "onfocus", "onblur", NULL } ;
676 static const char* const target_attr[] = { "target", NULL } ;
677 static const char* const rows_cols_attr[] = { "rows", "cols", NULL } ;
678 static const char* const alt_attr[] = { "alt", NULL } ;
679 static const char* const src_alt_attrs[] = { "src", "alt", NULL } ;
680 static const char* const href_attrs[] = { "href", NULL } ;
681 static const char* const clear_attrs[] = { "clear", NULL } ;
682 static const char* const inline_p[] = { INLINE, "p", NULL } ;
684 static const char* const flow_param[] = { FLOW, "param", NULL } ;
685 static const char* const applet_attrs[] = { COREATTRS , "codebase",
686 "archive", "alt", "name", "height", "width", "align",
687 "hspace", "vspace", NULL } ;
688 static const char* const area_attrs[] = { "shape", "coords", "href", "nohref",
689 "tabindex", "accesskey", "onfocus", "onblur", NULL } ;
690 static const char* const basefont_attrs[] =
691 { "id", "size", "color", "face", NULL } ;
692 static const char* const quote_attrs[] = { ATTRS, "cite", NULL } ;
693 static const char* const body_contents[] = { FLOW, "ins", "del", NULL } ;
694 static const char* const body_attrs[] = { ATTRS, "onload", "onunload", NULL } ;
695 static const char* const body_depr[] = { "background", "bgcolor", "text",
696 "link", "vlink", "alink", NULL } ;
697 static const char* const button_attrs[] = { ATTRS, "name", "value", "type",
698 "disabled", "tabindex", "accesskey", "onfocus", "onblur", NULL } ;
701 static const char* const col_attrs[] = { ATTRS, "span", "width", CELLHALIGN, CELLVALIGN, NULL } ;
702 static const char* const col_elt[] = { "col", NULL } ;
703 static const char* const edit_attrs[] = { ATTRS, "datetime", "cite", NULL } ;
704 static const char* const compact_attrs[] = { ATTRS, "compact", NULL } ;
705 static const char* const dl_contents[] = { "dt", "dd", NULL } ;
706 static const char* const compact_attr[] = { "compact", NULL } ;
707 static const char* const label_attr[] = { "label", NULL } ;
708 static const char* const fieldset_contents[] = { FLOW, "legend" } ;
709 static const char* const font_attrs[] = { COREATTRS, I18N, "size", "color", "face" , NULL } ;
710 static const char* const form_contents[] = { HEADING, LIST, INLINE, "pre", "p", "div", "center", "noscript", "noframes", "blockquote", "isindex", "hr", "table", "fieldset", "address", NULL } ;
711 static const char* const form_attrs[] = { ATTRS, "method", "enctype", "accept", "name", "onsubmit", "onreset", "accept-charset", NULL } ;
712 static const char* const frame_attrs[] = { COREATTRS, "longdesc", "name", "src", "frameborder", "marginwidth", "marginheight", "noresize", "scrolling" , NULL } ;
713 static const char* const frameset_attrs[] = { COREATTRS, "rows", "cols", "onload", "onunload", NULL } ;
714 static const char* const frameset_contents[] = { "frameset", "frame", "noframes", NULL } ;
715 static const char* const head_attrs[] = { I18N, "profile", NULL } ;
716 static const char* const head_contents[] = { "title", "isindex", "base", "script", "style", "meta", "link", "object", NULL } ;
717 static const char* const hr_depr[] = { "align", "noshade", "size", "width", NULL } ;
718 static const char* const version_attr[] = { "version", NULL } ;
719 static const char* const html_content[] = { "head", "body", "frameset", NULL } ;
720 static const char* const iframe_attrs[] = { COREATTRS, "longdesc", "name", "src", "frameborder", "marginwidth", "marginheight", "scrolling", "align", "height", "width", NULL } ;
721 static const char* const img_attrs[] = { ATTRS, "longdesc", "name", "height", "width", "usemap", "ismap", NULL } ;
722 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 } ;
723 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 } ;
724 static const char* const prompt_attrs[] = { COREATTRS, I18N, "prompt", NULL } ;
725 static const char* const label_attrs[] = { ATTRS, "for", "accesskey", "onfocus", "onblur", NULL } ;
726 static const char* const legend_attrs[] = { ATTRS, "accesskey", NULL } ;
727 static const char* const align_attr[] = { "align", NULL } ;
728 static const char* const link_attrs[] = { ATTRS, "charset", "href", "hreflang", "type", "rel", "rev", "media", NULL } ;
729 static const char* const map_contents[] = { BLOCK, "area", NULL } ;
730 static const char* const name_attr[] = { "name", NULL } ;
731 static const char* const action_attr[] = { "action", NULL } ;
732 static const char* const blockli_elt[] = { BLOCK, "li", NULL } ;
733 static const char* const meta_attrs[] = { I18N, "http-equiv", "name", "scheme", "charset", NULL } ;
734 static const char* const content_attr[] = { "content", NULL } ;
735 static const char* const type_attr[] = { "type", NULL } ;
736 static const char* const noframes_content[] = { "body", FLOW MODIFIER, NULL } ;
737 static const char* const object_contents[] = { FLOW, "param", NULL } ;
738 static const char* const object_attrs[] = { ATTRS, "declare", "classid", "codebase", "data", "type", "codetype", "archive", "standby", "height", "width", "usemap", "name", "tabindex", NULL } ;
739 static const char* const object_depr[] = { "align", "border", "hspace", "vspace", NULL } ;
740 static const char* const ol_attrs[] = { "type", "compact", "start", NULL} ;
741 static const char* const option_elt[] = { "option", NULL } ;
742 static const char* const optgroup_attrs[] = { ATTRS, "disabled", NULL } ;
743 static const char* const option_attrs[] = { ATTRS, "disabled", "label", "selected", "value", NULL } ;
744 static const char* const param_attrs[] = { "id", "value", "valuetype", "type", NULL } ;
745 static const char* const width_attr[] = { "width", NULL } ;
746 static const char* const pre_content[] = { PHRASE, "tt", "i", "b", "u", "s", "strike", "a", "br", "script", "map", "q", "span", "bdo", "iframe", NULL } ;
747 static const char* const script_attrs[] = { "charset", "src", "defer", "event", "for", NULL } ;
748 static const char* const language_attr[] = { "language", NULL } ;
749 static const char* const select_content[] = { "optgroup", "option", NULL } ;
750 static const char* const select_attrs[] = { ATTRS, "name", "size", "multiple", "disabled", "tabindex", "onfocus", "onblur", "onchange", NULL } ;
751 static const char* const style_attrs[] = { I18N, "media", "title", NULL } ;
752 static const char* const table_attrs[] = { ATTRS, "summary", "width", "border", "frame", "rules", "cellspacing", "cellpadding", "datapagesize", NULL } ;
753 static const char* const table_depr[] = { "align", "bgcolor", NULL } ;
754 static const char* const table_contents[] = { "caption", "col", "colgroup", "thead", "tfoot", "tbody", "tr", NULL} ;
755 static const char* const tr_elt[] = { "tr", NULL } ;
756 static const char* const talign_attrs[] = { ATTRS, CELLHALIGN, CELLVALIGN, NULL} ;
757 static const char* const th_td_depr[] = { "nowrap", "bgcolor", "width", "height", NULL } ;
758 static const char* const th_td_attr[] = { ATTRS, "abbr", "axis", "headers", "scope", "rowspan", "colspan", CELLHALIGN, CELLVALIGN, NULL } ;
759 static const char* const textarea_attrs[] = { ATTRS, "name", "disabled", "readonly", "tabindex", "accesskey", "onfocus", "onblur", "onselect", "onchange", NULL } ;
760 static const char* const tr_contents[] = { "th", "td", NULL } ;
761 static const char* const bgcolor_attr[] = { "bgcolor", NULL } ;
762 static const char* const li_elt[] = { "li", NULL } ;
763 static const char* const ul_depr[] = { "type", "compact", NULL} ;
764 static const char* const dir_attr[] = { "dir", NULL} ;
766 #define DECL (const char**)
768 static const htmlElemDesc
769 html40ElementTable[] = {
770 { "a", 0, 0, 0, 0, 0, 0, 1, "anchor ",
771 DECL html_inline , NULL , DECL a_attrs , DECL target_attr, NULL
773 { "abbr", 0, 0, 0, 0, 0, 0, 1, "abbreviated form",
774 DECL html_inline , NULL , DECL html_attrs, NULL, NULL
776 { "acronym", 0, 0, 0, 0, 0, 0, 1, "",
777 DECL html_inline , NULL , DECL html_attrs, NULL, NULL
779 { "address", 0, 0, 0, 0, 0, 0, 0, "information on author ",
780 DECL inline_p , NULL , DECL html_attrs, NULL, NULL
782 { "applet", 0, 0, 0, 0, 1, 1, 2, "java applet ",
783 DECL flow_param , NULL , NULL , DECL applet_attrs, NULL
785 { "area", 0, 2, 2, 1, 0, 0, 0, "client-side image map area ",
786 EMPTY , NULL , DECL area_attrs , DECL target_attr, DECL alt_attr
788 { "b", 0, 3, 0, 0, 0, 0, 1, "bold text style",
789 DECL html_inline , NULL , DECL html_attrs, NULL, NULL
791 { "base", 0, 2, 2, 1, 0, 0, 0, "document base uri ",
792 EMPTY , NULL , NULL , DECL target_attr, DECL href_attrs
794 { "basefont", 0, 2, 2, 1, 1, 1, 1, "base font size " ,
795 EMPTY , NULL , NULL, DECL basefont_attrs, NULL
797 { "bdo", 0, 0, 0, 0, 0, 0, 1, "i18n bidi over-ride ",
798 DECL html_inline , NULL , DECL core_i18n_attrs, NULL, DECL dir_attr
800 { "big", 0, 3, 0, 0, 0, 0, 1, "large text style",
801 DECL html_inline , NULL , DECL html_attrs, NULL, NULL
803 { "blockquote", 0, 0, 0, 0, 0, 0, 0, "long quotation ",
804 DECL html_flow , NULL , DECL quote_attrs , NULL, NULL
806 { "body", 1, 1, 0, 0, 0, 0, 0, "document body ",
807 DECL body_contents , "div" , DECL body_attrs, DECL body_depr, NULL
809 { "br", 0, 2, 2, 1, 0, 0, 1, "forced line break ",
810 EMPTY , NULL , DECL core_attrs, DECL clear_attrs , NULL
812 { "button", 0, 0, 0, 0, 0, 0, 2, "push button ",
813 DECL html_flow MODIFIER , NULL , DECL button_attrs, NULL, NULL
815 { "caption", 0, 0, 0, 0, 0, 0, 0, "table caption ",
816 DECL html_inline , NULL , DECL html_attrs, NULL, NULL
818 { "center", 0, 3, 0, 0, 1, 1, 0, "shorthand for div align=center ",
819 DECL html_flow , NULL , NULL, DECL html_attrs, NULL
821 { "cite", 0, 0, 0, 0, 0, 0, 1, "citation",
822 DECL html_inline , NULL , DECL html_attrs, NULL, NULL
824 { "code", 0, 0, 0, 0, 0, 0, 1, "computer code fragment",
825 DECL html_inline , NULL , DECL html_attrs, NULL, NULL
827 { "col", 0, 2, 2, 1, 0, 0, 0, "table column ",
828 EMPTY , NULL , DECL col_attrs , NULL, NULL
830 { "colgroup", 0, 1, 0, 0, 0, 0, 0, "table column group ",
831 DECL col_elt , "col" , DECL col_attrs , NULL, NULL
833 { "dd", 0, 1, 0, 0, 0, 0, 0, "definition description ",
834 DECL html_flow , NULL , DECL html_attrs, NULL, NULL
836 { "del", 0, 0, 0, 0, 0, 0, 2, "deleted text ",
837 DECL html_flow , NULL , DECL edit_attrs , NULL, NULL
839 { "dfn", 0, 0, 0, 0, 0, 0, 1, "instance definition",
840 DECL html_inline , NULL , DECL html_attrs, NULL, NULL
842 { "dir", 0, 0, 0, 0, 1, 1, 0, "directory list",
843 DECL blockli_elt, "li" , NULL, DECL compact_attrs, NULL
845 { "div", 0, 0, 0, 0, 0, 0, 0, "generic language/style container",
846 DECL html_flow, NULL, DECL html_attrs, DECL align_attr, NULL
848 { "dl", 0, 0, 0, 0, 0, 0, 0, "definition list ",
849 DECL dl_contents , "dd" , DECL html_attrs, DECL compact_attr, NULL
851 { "dt", 0, 1, 0, 0, 0, 0, 0, "definition term ",
852 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
854 { "em", 0, 3, 0, 0, 0, 0, 1, "emphasis",
855 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
857 { "embed", 0, 1, 0, 0, 1, 1, 1, "generic embedded object ",
858 EMPTY, NULL, DECL embed_attrs, NULL, NULL
860 { "fieldset", 0, 0, 0, 0, 0, 0, 0, "form control group ",
861 DECL fieldset_contents , NULL, DECL html_attrs, NULL, NULL
863 { "font", 0, 3, 0, 0, 1, 1, 1, "local change to font ",
864 DECL html_inline, NULL, NULL, DECL font_attrs, NULL
866 { "form", 0, 0, 0, 0, 0, 0, 0, "interactive form ",
867 DECL form_contents, "fieldset", DECL form_attrs , DECL target_attr, DECL action_attr
869 { "frame", 0, 2, 2, 1, 0, 2, 0, "subwindow " ,
870 EMPTY, NULL, NULL, DECL frame_attrs, NULL
872 { "frameset", 0, 0, 0, 0, 0, 2, 0, "window subdivision" ,
873 DECL frameset_contents, "noframes" , NULL , DECL frameset_attrs, NULL
875 { "h1", 0, 0, 0, 0, 0, 0, 0, "heading ",
876 DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
878 { "h2", 0, 0, 0, 0, 0, 0, 0, "heading ",
879 DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
881 { "h3", 0, 0, 0, 0, 0, 0, 0, "heading ",
882 DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
884 { "h4", 0, 0, 0, 0, 0, 0, 0, "heading ",
885 DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
887 { "h5", 0, 0, 0, 0, 0, 0, 0, "heading ",
888 DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
890 { "h6", 0, 0, 0, 0, 0, 0, 0, "heading ",
891 DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
893 { "head", 1, 1, 0, 0, 0, 0, 0, "document head ",
894 DECL head_contents, NULL, DECL head_attrs, NULL, NULL
896 { "hr", 0, 2, 2, 1, 0, 0, 0, "horizontal rule " ,
897 EMPTY, NULL, DECL html_attrs, DECL hr_depr, NULL
899 { "html", 1, 1, 0, 0, 0, 0, 0, "document root element ",
900 DECL html_content , NULL , DECL i18n_attrs, DECL version_attr, NULL
902 { "i", 0, 3, 0, 0, 0, 0, 1, "italic text style",
903 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
905 { "iframe", 0, 0, 0, 0, 0, 1, 2, "inline subwindow ",
906 DECL html_flow, NULL, NULL, DECL iframe_attrs, NULL
908 { "img", 0, 2, 2, 1, 0, 0, 1, "embedded image ",
909 EMPTY, NULL, DECL img_attrs, DECL align_attr, DECL src_alt_attrs
911 { "input", 0, 2, 2, 1, 0, 0, 1, "form control ",
912 EMPTY, NULL, DECL input_attrs , DECL align_attr, NULL
914 { "ins", 0, 0, 0, 0, 0, 0, 2, "inserted text",
915 DECL html_flow, NULL, DECL edit_attrs, NULL, NULL
917 { "isindex", 0, 2, 2, 1, 1, 1, 0, "single line prompt ",
918 EMPTY, NULL, NULL, DECL prompt_attrs, NULL
920 { "kbd", 0, 0, 0, 0, 0, 0, 1, "text to be entered by the user",
921 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
923 { "label", 0, 0, 0, 0, 0, 0, 1, "form field label text ",
924 DECL html_inline MODIFIER, NULL, DECL label_attrs , NULL, NULL
926 { "legend", 0, 0, 0, 0, 0, 0, 0, "fieldset legend ",
927 DECL html_inline, NULL, DECL legend_attrs , DECL align_attr, NULL
929 { "li", 0, 1, 1, 0, 0, 0, 0, "list item ",
930 DECL html_flow, NULL, DECL html_attrs, NULL, NULL
932 { "link", 0, 2, 2, 1, 0, 0, 0, "a media-independent link ",
933 EMPTY, NULL, DECL link_attrs, DECL target_attr, NULL
935 { "map", 0, 0, 0, 0, 0, 0, 2, "client-side image map ",
936 DECL map_contents , NULL, DECL html_attrs , NULL, DECL name_attr
938 { "menu", 0, 0, 0, 0, 1, 1, 0, "menu list ",
939 DECL blockli_elt , NULL, NULL, DECL compact_attrs, NULL
941 { "meta", 0, 2, 2, 1, 0, 0, 0, "generic metainformation ",
942 EMPTY, NULL, DECL meta_attrs , NULL , DECL content_attr
944 { "noframes", 0, 0, 0, 0, 0, 2, 0, "alternate content container for non frame-based rendering ",
945 DECL noframes_content, "body" , DECL html_attrs, NULL, NULL
947 { "noscript", 0, 0, 0, 0, 0, 0, 0, "alternate content container for non script-based rendering ",
948 DECL html_flow, "div", DECL html_attrs, NULL, NULL
950 { "object", 0, 0, 0, 0, 0, 0, 2, "generic embedded object ",
951 DECL object_contents , "div" , DECL object_attrs, DECL object_depr, NULL
953 { "ol", 0, 0, 0, 0, 0, 0, 0, "ordered list ",
954 DECL li_elt , "li" , DECL html_attrs, DECL ol_attrs, NULL
956 { "optgroup", 0, 0, 0, 0, 0, 0, 0, "option group ",
957 DECL option_elt , "option", DECL optgroup_attrs, NULL, DECL label_attr
959 { "option", 0, 1, 0, 0, 0, 0, 0, "selectable choice " ,
960 DECL html_pcdata, NULL, DECL option_attrs, NULL, NULL
962 { "p", 0, 1, 0, 0, 0, 0, 0, "paragraph ",
963 DECL html_inline, NULL, DECL html_attrs, DECL align_attr, NULL
965 { "param", 0, 2, 2, 1, 0, 0, 0, "named property value ",
966 EMPTY, NULL, DECL param_attrs, NULL, DECL name_attr
968 { "pre", 0, 0, 0, 0, 0, 0, 0, "preformatted text ",
969 DECL pre_content, NULL, DECL html_attrs, DECL width_attr, NULL
971 { "q", 0, 0, 0, 0, 0, 0, 1, "short inline quotation ",
972 DECL html_inline, NULL, DECL quote_attrs, NULL, NULL
974 { "s", 0, 3, 0, 0, 1, 1, 1, "strike-through text style",
975 DECL html_inline, NULL, NULL, DECL html_attrs, NULL
977 { "samp", 0, 0, 0, 0, 0, 0, 1, "sample program output, scripts, etc.",
978 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
980 { "script", 0, 0, 0, 0, 0, 0, 2, "script statements ",
981 DECL html_cdata, NULL, DECL script_attrs, DECL language_attr, DECL type_attr
983 { "select", 0, 0, 0, 0, 0, 0, 1, "option selector ",
984 DECL select_content, NULL, DECL select_attrs, NULL, NULL
986 { "small", 0, 3, 0, 0, 0, 0, 1, "small text style",
987 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
989 { "span", 0, 0, 0, 0, 0, 0, 1, "generic language/style container ",
990 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
992 { "strike", 0, 3, 0, 0, 1, 1, 1, "strike-through text",
993 DECL html_inline, NULL, NULL, DECL html_attrs, NULL
995 { "strong", 0, 3, 0, 0, 0, 0, 1, "strong emphasis",
996 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
998 { "style", 0, 0, 0, 0, 0, 0, 0, "style info ",
999 DECL html_cdata, NULL, DECL style_attrs, NULL, DECL type_attr
1001 { "sub", 0, 3, 0, 0, 0, 0, 1, "subscript",
1002 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
1004 { "sup", 0, 3, 0, 0, 0, 0, 1, "superscript ",
1005 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
1007 { "table", 0, 0, 0, 0, 0, 0, 0, "",
1008 DECL table_contents , "tr" , DECL table_attrs , DECL table_depr, NULL
1010 { "tbody", 1, 0, 0, 0, 0, 0, 0, "table body ",
1011 DECL tr_elt , "tr" , DECL talign_attrs, NULL, NULL
1013 { "td", 0, 0, 0, 0, 0, 0, 0, "table data cell",
1014 DECL html_flow, NULL, DECL th_td_attr, DECL th_td_depr, NULL
1016 { "textarea", 0, 0, 0, 0, 0, 0, 1, "multi-line text field ",
1017 DECL html_pcdata, NULL, DECL textarea_attrs, NULL, DECL rows_cols_attr
1019 { "tfoot", 0, 1, 0, 0, 0, 0, 0, "table footer ",
1020 DECL tr_elt , "tr" , DECL talign_attrs, NULL, NULL
1022 { "th", 0, 1, 0, 0, 0, 0, 0, "table header cell",
1023 DECL html_flow, NULL, DECL th_td_attr, DECL th_td_depr, NULL
1025 { "thead", 0, 1, 0, 0, 0, 0, 0, "table header ",
1026 DECL tr_elt , "tr" , DECL talign_attrs, NULL, NULL
1028 { "title", 0, 0, 0, 0, 0, 0, 0, "document title ",
1029 DECL html_pcdata, NULL, DECL i18n_attrs, NULL, NULL
1031 { "tr", 0, 0, 0, 0, 0, 0, 0, "table row ",
1032 DECL tr_contents , "td" , DECL talign_attrs, DECL bgcolor_attr, NULL
1034 { "tt", 0, 3, 0, 0, 0, 0, 1, "teletype or monospaced text style",
1035 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
1037 { "u", 0, 3, 0, 0, 1, 1, 1, "underlined text style",
1038 DECL html_inline, NULL, NULL, DECL html_attrs, NULL
1040 { "ul", 0, 0, 0, 0, 0, 0, 0, "unordered list ",
1041 DECL li_elt , "li" , DECL html_attrs, DECL ul_depr, NULL
1043 { "var", 0, 0, 0, 0, 0, 0, 1, "instance of a variable or program argument",
1044 DECL html_inline, NULL, DECL html_attrs, NULL, NULL
1049 * start tags that imply the end of current element
1051 static const char * const htmlStartClose[] = {
1052 "form", "form", "p", "hr", "h1", "h2", "h3", "h4", "h5", "h6",
1053 "dl", "ul", "ol", "menu", "dir", "address", "pre",
1054 "listing", "xmp", "head", NULL,
1057 "body", "head", "style", "link", "title", "p", NULL,
1058 "frameset", "head", "style", "link", "title", "p", NULL,
1059 "li", "p", "h1", "h2", "h3", "h4", "h5", "h6", "dl", "address",
1060 "pre", "listing", "xmp", "head", "li", NULL,
1061 "hr", "p", "head", NULL,
1062 "h1", "p", "head", NULL,
1063 "h2", "p", "head", NULL,
1064 "h3", "p", "head", NULL,
1065 "h4", "p", "head", NULL,
1066 "h5", "p", "head", NULL,
1067 "h6", "p", "head", NULL,
1068 "dir", "p", "head", NULL,
1069 "address", "p", "head", "ul", NULL,
1070 "pre", "p", "head", "ul", NULL,
1071 "listing", "p", "head", NULL,
1072 "xmp", "p", "head", NULL,
1073 "blockquote", "p", "head", NULL,
1074 "dl", "p", "dt", "menu", "dir", "address", "pre", "listing",
1075 "xmp", "head", NULL,
1076 "dt", "p", "menu", "dir", "address", "pre", "listing", "xmp",
1078 "dd", "p", "menu", "dir", "address", "pre", "listing", "xmp",
1080 "ul", "p", "head", "ol", "menu", "dir", "address", "pre",
1081 "listing", "xmp", NULL,
1082 "ol", "p", "head", "ul", NULL,
1083 "menu", "p", "head", "ul", NULL,
1084 "p", "p", "head", "h1", "h2", "h3", "h4", "h5", "h6", FONTSTYLE, NULL,
1085 "div", "p", "head", NULL,
1086 "noscript", "p", NULL,
1087 "center", "font", "b", "i", "p", "head", NULL,
1088 "a", "a", "head", NULL,
1089 "caption", "p", NULL,
1090 "colgroup", "caption", "colgroup", "col", "p", NULL,
1091 "col", "caption", "col", "p", NULL,
1092 "table", "p", "head", "h1", "h2", "h3", "h4", "h5", "h6", "pre",
1093 "listing", "xmp", "a", NULL,
1094 "th", "th", "td", "p", "span", "font", "a", "b", "i", "u", NULL,
1095 "td", "th", "td", "p", "span", "font", "a", "b", "i", "u", NULL,
1096 "tr", "th", "td", "tr", "caption", "col", "colgroup", "p", NULL,
1097 "thead", "caption", "col", "colgroup", NULL,
1098 "tfoot", "th", "td", "tr", "caption", "col", "colgroup", "thead",
1100 "tbody", "th", "td", "tr", "caption", "col", "colgroup", "thead",
1101 "tfoot", "tbody", "p", NULL,
1102 "optgroup", "option", NULL,
1103 "option", "option", NULL,
1104 "fieldset", "legend", "p", "head", "h1", "h2", "h3", "h4", "h5", "h6",
1105 "pre", "listing", "xmp", "a", NULL,
1106 /* most tags in in FONTSTYLE, PHRASE and SPECIAL should close <head> */
1112 "strike", "head", NULL,
1113 "big", "head", NULL,
1114 "small", "head", NULL,
1117 "strong", "head", NULL,
1118 "dfn", "head", NULL,
1119 "code", "head", NULL,
1120 "samp", "head", NULL,
1121 "kbd", "head", NULL,
1122 "var", "head", NULL,
1123 "cite", "head", NULL,
1124 "abbr", "head", NULL,
1125 "acronym", "head", NULL,
1128 "img", "head", NULL,
1132 "font", "head", NULL,
1136 "map", "head", NULL,
1138 "sub", "head", NULL,
1139 "sup", "head", NULL,
1140 "span", "head", NULL,
1141 "bdo", "head", NULL,
1142 "iframe", "head", NULL,
1147 * The list of HTML elements which are supposed not to have
1148 * CDATA content and where a p element will be implied
1150 * TODO: extend that list by reading the HTML SGML DTD on
1153 static const char *const htmlNoContentElements[] = {
1160 * The list of HTML attributes which are of content %Script;
1161 * NOTE: when adding ones, check htmlIsScriptAttribute() since
1162 * it assumes the name starts with 'on'
1164 static const char *const htmlScriptAttributes[] = {
1186 * This table is used by the htmlparser to know what to do with
1187 * broken html pages. By assigning different priorities to different
1188 * elements the parser can decide how to handle extra endtags.
1189 * Endtags are only allowed to close elements with lower or equal
1198 static const elementPriority htmlEndPriority[] = {
1210 {NULL, 100} /* Default priority */
1213 static const char** htmlStartCloseIndex[100];
1214 static int htmlStartCloseIndexinitialized = 0;
1216 /************************************************************************
1218 * functions to handle HTML specific data *
1220 ************************************************************************/
1223 * htmlInitAutoClose:
1225 * Initialize the htmlStartCloseIndex for fast lookup of closing tags names.
1226 * This is not reentrant. Call xmlInitParser() once before processing in
1227 * case of use in multithreaded programs.
1230 htmlInitAutoClose(void) {
1233 if (htmlStartCloseIndexinitialized) return;
1235 for (indx = 0;indx < 100;indx ++) htmlStartCloseIndex[indx] = NULL;
1237 while ((htmlStartClose[i] != NULL) && (indx < 100 - 1)) {
1238 htmlStartCloseIndex[indx++] = (const char**) &htmlStartClose[i];
1239 while (htmlStartClose[i] != NULL) i++;
1242 htmlStartCloseIndexinitialized = 1;
1247 * @tag: The tag name in lowercase
1249 * Lookup the HTML tag in the ElementTable
1251 * Returns the related htmlElemDescPtr or NULL if not found.
1253 const htmlElemDesc *
1254 htmlTagLookup(const xmlChar *tag) {
1257 for (i = 0; i < (sizeof(html40ElementTable) /
1258 sizeof(html40ElementTable[0]));i++) {
1259 if (!xmlStrcasecmp(tag, BAD_CAST html40ElementTable[i].name))
1260 return((htmlElemDescPtr) &html40ElementTable[i]);
1266 * htmlGetEndPriority:
1267 * @name: The name of the element to look up the priority for.
1269 * Return value: The "endtag" priority.
1272 htmlGetEndPriority (const xmlChar *name) {
1275 while ((htmlEndPriority[i].name != NULL) &&
1276 (!xmlStrEqual((const xmlChar *)htmlEndPriority[i].name, name)))
1279 return(htmlEndPriority[i].priority);
1284 * htmlCheckAutoClose:
1285 * @newtag: The new tag name
1286 * @oldtag: The old tag name
1288 * Checks whether the new tag is one of the registered valid tags for
1290 * Initialize the htmlStartCloseIndex for fast lookup of closing tags names.
1292 * Returns 0 if no, 1 if yes.
1295 htmlCheckAutoClose(const xmlChar * newtag, const xmlChar * oldtag)
1298 const char **closed = NULL;
1300 if (htmlStartCloseIndexinitialized == 0)
1301 htmlInitAutoClose();
1303 /* inefficient, but not a big deal */
1304 for (indx = 0; indx < 100; indx++) {
1305 closed = htmlStartCloseIndex[indx];
1308 if (xmlStrEqual(BAD_CAST * closed, newtag))
1312 i = closed - htmlStartClose;
1314 while (htmlStartClose[i] != NULL) {
1315 if (xmlStrEqual(BAD_CAST htmlStartClose[i], oldtag)) {
1324 * htmlAutoCloseOnClose:
1325 * @ctxt: an HTML parser context
1326 * @newtag: The new tag name
1327 * @force: force the tag closure
1329 * The HTML DTD allows an ending tag to implicitly close other tags.
1332 htmlAutoCloseOnClose(htmlParserCtxtPtr ctxt, const xmlChar * newtag)
1334 const htmlElemDesc *info;
1337 priority = htmlGetEndPriority(newtag);
1339 for (i = (ctxt->nameNr - 1); i >= 0; i--) {
1341 if (xmlStrEqual(newtag, ctxt->nameTab[i]))
1344 * A missplaced endtag can only close elements with lower
1345 * or equal priority, so if we find an element with higher
1346 * priority before we find an element with
1347 * matching name, we just ignore this endtag
1349 if (htmlGetEndPriority(ctxt->nameTab[i]) > priority)
1355 while (!xmlStrEqual(newtag, ctxt->name)) {
1356 info = htmlTagLookup(ctxt->name);
1357 if ((info != NULL) && (info->endTag == 3)) {
1358 htmlParseErr(ctxt, XML_ERR_TAG_NAME_MISMATCH,
1359 "Opening and ending tag mismatch: %s and %s\n",
1360 newtag, ctxt->name);
1362 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
1363 ctxt->sax->endElement(ctxt->userData, ctxt->name);
1369 * htmlAutoCloseOnEnd:
1370 * @ctxt: an HTML parser context
1372 * Close all remaining tags at the end of the stream
1375 htmlAutoCloseOnEnd(htmlParserCtxtPtr ctxt)
1379 if (ctxt->nameNr == 0)
1381 for (i = (ctxt->nameNr - 1); i >= 0; i--) {
1382 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
1383 ctxt->sax->endElement(ctxt->userData, ctxt->name);
1390 * @ctxt: an HTML parser context
1391 * @newtag: The new tag name or NULL
1393 * The HTML DTD allows a tag to implicitly close other tags.
1394 * The list is kept in htmlStartClose array. This function is
1395 * called when a new tag has been detected and generates the
1396 * appropriates closes if possible/needed.
1397 * If newtag is NULL this mean we are at the end of the resource
1398 * and we should check
1401 htmlAutoClose(htmlParserCtxtPtr ctxt, const xmlChar * newtag)
1403 while ((newtag != NULL) && (ctxt->name != NULL) &&
1404 (htmlCheckAutoClose(newtag, ctxt->name))) {
1405 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
1406 ctxt->sax->endElement(ctxt->userData, ctxt->name);
1409 if (newtag == NULL) {
1410 htmlAutoCloseOnEnd(ctxt);
1413 while ((newtag == NULL) && (ctxt->name != NULL) &&
1414 ((xmlStrEqual(ctxt->name, BAD_CAST "head")) ||
1415 (xmlStrEqual(ctxt->name, BAD_CAST "body")) ||
1416 (xmlStrEqual(ctxt->name, BAD_CAST "html")))) {
1417 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
1418 ctxt->sax->endElement(ctxt->userData, ctxt->name);
1425 * @doc: the HTML document
1426 * @name: The tag name
1427 * @elem: the HTML element
1429 * The HTML DTD allows a tag to implicitly close other tags.
1430 * The list is kept in htmlStartClose array. This function checks
1431 * if the element or one of it's children would autoclose the
1434 * Returns 1 if autoclose, 0 otherwise
1437 htmlAutoCloseTag(htmlDocPtr doc, const xmlChar *name, htmlNodePtr elem) {
1440 if (elem == NULL) return(1);
1441 if (xmlStrEqual(name, elem->name)) return(0);
1442 if (htmlCheckAutoClose(elem->name, name)) return(1);
1443 child = elem->children;
1444 while (child != NULL) {
1445 if (htmlAutoCloseTag(doc, name, child)) return(1);
1446 child = child->next;
1453 * @doc: the HTML document
1454 * @elem: the HTML element
1456 * The HTML DTD allows a tag to implicitly close other tags.
1457 * The list is kept in htmlStartClose array. This function checks
1458 * if a tag is autoclosed by one of it's child
1460 * Returns 1 if autoclosed, 0 otherwise
1463 htmlIsAutoClosed(htmlDocPtr doc, htmlNodePtr elem) {
1466 if (elem == NULL) return(1);
1467 child = elem->children;
1468 while (child != NULL) {
1469 if (htmlAutoCloseTag(doc, elem->name, child)) return(1);
1470 child = child->next;
1477 * @ctxt: an HTML parser context
1478 * @newtag: The new tag name
1480 * The HTML DTD allows a tag to exists only implicitly
1481 * called when a new tag has been detected and generates the
1482 * appropriates implicit tags if missing
1485 htmlCheckImplied(htmlParserCtxtPtr ctxt, const xmlChar *newtag) {
1488 if (ctxt->options & HTML_PARSE_NOIMPLIED)
1490 if (!htmlOmittedDefaultValue)
1492 if (xmlStrEqual(newtag, BAD_CAST"html"))
1494 if (ctxt->nameNr <= 0) {
1495 htmlnamePush(ctxt, BAD_CAST"html");
1496 if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
1497 ctxt->sax->startElement(ctxt->userData, BAD_CAST"html", NULL);
1499 if ((xmlStrEqual(newtag, BAD_CAST"body")) || (xmlStrEqual(newtag, BAD_CAST"head")))
1501 if ((ctxt->nameNr <= 1) &&
1502 ((xmlStrEqual(newtag, BAD_CAST"script")) ||
1503 (xmlStrEqual(newtag, BAD_CAST"style")) ||
1504 (xmlStrEqual(newtag, BAD_CAST"meta")) ||
1505 (xmlStrEqual(newtag, BAD_CAST"link")) ||
1506 (xmlStrEqual(newtag, BAD_CAST"title")) ||
1507 (xmlStrEqual(newtag, BAD_CAST"base")))) {
1508 if (ctxt->html >= 3) {
1509 /* we already saw or generated an <head> before */
1513 * dropped OBJECT ... i you put it first BODY will be
1516 htmlnamePush(ctxt, BAD_CAST"head");
1517 if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
1518 ctxt->sax->startElement(ctxt->userData, BAD_CAST"head", NULL);
1519 } else if ((!xmlStrEqual(newtag, BAD_CAST"noframes")) &&
1520 (!xmlStrEqual(newtag, BAD_CAST"frame")) &&
1521 (!xmlStrEqual(newtag, BAD_CAST"frameset"))) {
1522 if (ctxt->html >= 10) {
1523 /* we already saw or generated a <body> before */
1526 for (i = 0;i < ctxt->nameNr;i++) {
1527 if (xmlStrEqual(ctxt->nameTab[i], BAD_CAST"body")) {
1530 if (xmlStrEqual(ctxt->nameTab[i], BAD_CAST"head")) {
1535 htmlnamePush(ctxt, BAD_CAST"body");
1536 if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
1537 ctxt->sax->startElement(ctxt->userData, BAD_CAST"body", NULL);
1542 * htmlCheckParagraph
1543 * @ctxt: an HTML parser context
1545 * Check whether a p element need to be implied before inserting
1546 * characters in the current element.
1548 * Returns 1 if a paragraph has been inserted, 0 if not and -1
1553 htmlCheckParagraph(htmlParserCtxtPtr ctxt) {
1561 htmlAutoClose(ctxt, BAD_CAST"p");
1562 htmlCheckImplied(ctxt, BAD_CAST"p");
1563 htmlnamePush(ctxt, BAD_CAST"p");
1564 if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
1565 ctxt->sax->startElement(ctxt->userData, BAD_CAST"p", NULL);
1568 if (!htmlOmittedDefaultValue)
1570 for (i = 0; htmlNoContentElements[i] != NULL; i++) {
1571 if (xmlStrEqual(tag, BAD_CAST htmlNoContentElements[i])) {
1572 htmlAutoClose(ctxt, BAD_CAST"p");
1573 htmlCheckImplied(ctxt, BAD_CAST"p");
1574 htmlnamePush(ctxt, BAD_CAST"p");
1575 if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
1576 ctxt->sax->startElement(ctxt->userData, BAD_CAST"p", NULL);
1584 * htmlIsScriptAttribute:
1585 * @name: an attribute name
1587 * Check if an attribute is of content type Script
1589 * Returns 1 is the attribute is a script 0 otherwise
1592 htmlIsScriptAttribute(const xmlChar *name) {
1598 * all script attributes start with 'on'
1600 if ((name[0] != 'o') || (name[1] != 'n'))
1603 i < sizeof(htmlScriptAttributes)/sizeof(htmlScriptAttributes[0]);
1605 if (xmlStrEqual(name, (const xmlChar *) htmlScriptAttributes[i]))
1611 /************************************************************************
1613 * The list of HTML predefined entities *
1615 ************************************************************************/
1618 static const htmlEntityDesc html40EntitiesTable[] = {
1620 * the 4 absolute ones, plus apostrophe.
1622 { 34, "quot", "quotation mark = APL quote, U+0022 ISOnum" },
1623 { 38, "amp", "ampersand, U+0026 ISOnum" },
1624 { 39, "apos", "single quote" },
1625 { 60, "lt", "less-than sign, U+003C ISOnum" },
1626 { 62, "gt", "greater-than sign, U+003E ISOnum" },
1629 * A bunch still in the 128-255 range
1630 * Replacing them depend really on the charset used.
1632 { 160, "nbsp", "no-break space = non-breaking space, U+00A0 ISOnum" },
1633 { 161, "iexcl","inverted exclamation mark, U+00A1 ISOnum" },
1634 { 162, "cent", "cent sign, U+00A2 ISOnum" },
1635 { 163, "pound","pound sign, U+00A3 ISOnum" },
1636 { 164, "curren","currency sign, U+00A4 ISOnum" },
1637 { 165, "yen", "yen sign = yuan sign, U+00A5 ISOnum" },
1638 { 166, "brvbar","broken bar = broken vertical bar, U+00A6 ISOnum" },
1639 { 167, "sect", "section sign, U+00A7 ISOnum" },
1640 { 168, "uml", "diaeresis = spacing diaeresis, U+00A8 ISOdia" },
1641 { 169, "copy", "copyright sign, U+00A9 ISOnum" },
1642 { 170, "ordf", "feminine ordinal indicator, U+00AA ISOnum" },
1643 { 171, "laquo","left-pointing double angle quotation mark = left pointing guillemet, U+00AB ISOnum" },
1644 { 172, "not", "not sign, U+00AC ISOnum" },
1645 { 173, "shy", "soft hyphen = discretionary hyphen, U+00AD ISOnum" },
1646 { 174, "reg", "registered sign = registered trade mark sign, U+00AE ISOnum" },
1647 { 175, "macr", "macron = spacing macron = overline = APL overbar, U+00AF ISOdia" },
1648 { 176, "deg", "degree sign, U+00B0 ISOnum" },
1649 { 177, "plusmn","plus-minus sign = plus-or-minus sign, U+00B1 ISOnum" },
1650 { 178, "sup2", "superscript two = superscript digit two = squared, U+00B2 ISOnum" },
1651 { 179, "sup3", "superscript three = superscript digit three = cubed, U+00B3 ISOnum" },
1652 { 180, "acute","acute accent = spacing acute, U+00B4 ISOdia" },
1653 { 181, "micro","micro sign, U+00B5 ISOnum" },
1654 { 182, "para", "pilcrow sign = paragraph sign, U+00B6 ISOnum" },
1655 { 183, "middot","middle dot = Georgian comma Greek middle dot, U+00B7 ISOnum" },
1656 { 184, "cedil","cedilla = spacing cedilla, U+00B8 ISOdia" },
1657 { 185, "sup1", "superscript one = superscript digit one, U+00B9 ISOnum" },
1658 { 186, "ordm", "masculine ordinal indicator, U+00BA ISOnum" },
1659 { 187, "raquo","right-pointing double angle quotation mark right pointing guillemet, U+00BB ISOnum" },
1660 { 188, "frac14","vulgar fraction one quarter = fraction one quarter, U+00BC ISOnum" },
1661 { 189, "frac12","vulgar fraction one half = fraction one half, U+00BD ISOnum" },
1662 { 190, "frac34","vulgar fraction three quarters = fraction three quarters, U+00BE ISOnum" },
1663 { 191, "iquest","inverted question mark = turned question mark, U+00BF ISOnum" },
1664 { 192, "Agrave","latin capital letter A with grave = latin capital letter A grave, U+00C0 ISOlat1" },
1665 { 193, "Aacute","latin capital letter A with acute, U+00C1 ISOlat1" },
1666 { 194, "Acirc","latin capital letter A with circumflex, U+00C2 ISOlat1" },
1667 { 195, "Atilde","latin capital letter A with tilde, U+00C3 ISOlat1" },
1668 { 196, "Auml", "latin capital letter A with diaeresis, U+00C4 ISOlat1" },
1669 { 197, "Aring","latin capital letter A with ring above = latin capital letter A ring, U+00C5 ISOlat1" },
1670 { 198, "AElig","latin capital letter AE = latin capital ligature AE, U+00C6 ISOlat1" },
1671 { 199, "Ccedil","latin capital letter C with cedilla, U+00C7 ISOlat1" },
1672 { 200, "Egrave","latin capital letter E with grave, U+00C8 ISOlat1" },
1673 { 201, "Eacute","latin capital letter E with acute, U+00C9 ISOlat1" },
1674 { 202, "Ecirc","latin capital letter E with circumflex, U+00CA ISOlat1" },
1675 { 203, "Euml", "latin capital letter E with diaeresis, U+00CB ISOlat1" },
1676 { 204, "Igrave","latin capital letter I with grave, U+00CC ISOlat1" },
1677 { 205, "Iacute","latin capital letter I with acute, U+00CD ISOlat1" },
1678 { 206, "Icirc","latin capital letter I with circumflex, U+00CE ISOlat1" },
1679 { 207, "Iuml", "latin capital letter I with diaeresis, U+00CF ISOlat1" },
1680 { 208, "ETH", "latin capital letter ETH, U+00D0 ISOlat1" },
1681 { 209, "Ntilde","latin capital letter N with tilde, U+00D1 ISOlat1" },
1682 { 210, "Ograve","latin capital letter O with grave, U+00D2 ISOlat1" },
1683 { 211, "Oacute","latin capital letter O with acute, U+00D3 ISOlat1" },
1684 { 212, "Ocirc","latin capital letter O with circumflex, U+00D4 ISOlat1" },
1685 { 213, "Otilde","latin capital letter O with tilde, U+00D5 ISOlat1" },
1686 { 214, "Ouml", "latin capital letter O with diaeresis, U+00D6 ISOlat1" },
1687 { 215, "times","multiplication sign, U+00D7 ISOnum" },
1688 { 216, "Oslash","latin capital letter O with stroke latin capital letter O slash, U+00D8 ISOlat1" },
1689 { 217, "Ugrave","latin capital letter U with grave, U+00D9 ISOlat1" },
1690 { 218, "Uacute","latin capital letter U with acute, U+00DA ISOlat1" },
1691 { 219, "Ucirc","latin capital letter U with circumflex, U+00DB ISOlat1" },
1692 { 220, "Uuml", "latin capital letter U with diaeresis, U+00DC ISOlat1" },
1693 { 221, "Yacute","latin capital letter Y with acute, U+00DD ISOlat1" },
1694 { 222, "THORN","latin capital letter THORN, U+00DE ISOlat1" },
1695 { 223, "szlig","latin small letter sharp s = ess-zed, U+00DF ISOlat1" },
1696 { 224, "agrave","latin small letter a with grave = latin small letter a grave, U+00E0 ISOlat1" },
1697 { 225, "aacute","latin small letter a with acute, U+00E1 ISOlat1" },
1698 { 226, "acirc","latin small letter a with circumflex, U+00E2 ISOlat1" },
1699 { 227, "atilde","latin small letter a with tilde, U+00E3 ISOlat1" },
1700 { 228, "auml", "latin small letter a with diaeresis, U+00E4 ISOlat1" },
1701 { 229, "aring","latin small letter a with ring above = latin small letter a ring, U+00E5 ISOlat1" },
1702 { 230, "aelig","latin small letter ae = latin small ligature ae, U+00E6 ISOlat1" },
1703 { 231, "ccedil","latin small letter c with cedilla, U+00E7 ISOlat1" },
1704 { 232, "egrave","latin small letter e with grave, U+00E8 ISOlat1" },
1705 { 233, "eacute","latin small letter e with acute, U+00E9 ISOlat1" },
1706 { 234, "ecirc","latin small letter e with circumflex, U+00EA ISOlat1" },
1707 { 235, "euml", "latin small letter e with diaeresis, U+00EB ISOlat1" },
1708 { 236, "igrave","latin small letter i with grave, U+00EC ISOlat1" },
1709 { 237, "iacute","latin small letter i with acute, U+00ED ISOlat1" },
1710 { 238, "icirc","latin small letter i with circumflex, U+00EE ISOlat1" },
1711 { 239, "iuml", "latin small letter i with diaeresis, U+00EF ISOlat1" },
1712 { 240, "eth", "latin small letter eth, U+00F0 ISOlat1" },
1713 { 241, "ntilde","latin small letter n with tilde, U+00F1 ISOlat1" },
1714 { 242, "ograve","latin small letter o with grave, U+00F2 ISOlat1" },
1715 { 243, "oacute","latin small letter o with acute, U+00F3 ISOlat1" },
1716 { 244, "ocirc","latin small letter o with circumflex, U+00F4 ISOlat1" },
1717 { 245, "otilde","latin small letter o with tilde, U+00F5 ISOlat1" },
1718 { 246, "ouml", "latin small letter o with diaeresis, U+00F6 ISOlat1" },
1719 { 247, "divide","division sign, U+00F7 ISOnum" },
1720 { 248, "oslash","latin small letter o with stroke, = latin small letter o slash, U+00F8 ISOlat1" },
1721 { 249, "ugrave","latin small letter u with grave, U+00F9 ISOlat1" },
1722 { 250, "uacute","latin small letter u with acute, U+00FA ISOlat1" },
1723 { 251, "ucirc","latin small letter u with circumflex, U+00FB ISOlat1" },
1724 { 252, "uuml", "latin small letter u with diaeresis, U+00FC ISOlat1" },
1725 { 253, "yacute","latin small letter y with acute, U+00FD ISOlat1" },
1726 { 254, "thorn","latin small letter thorn with, U+00FE ISOlat1" },
1727 { 255, "yuml", "latin small letter y with diaeresis, U+00FF ISOlat1" },
1729 { 338, "OElig","latin capital ligature OE, U+0152 ISOlat2" },
1730 { 339, "oelig","latin small ligature oe, U+0153 ISOlat2" },
1731 { 352, "Scaron","latin capital letter S with caron, U+0160 ISOlat2" },
1732 { 353, "scaron","latin small letter s with caron, U+0161 ISOlat2" },
1733 { 376, "Yuml", "latin capital letter Y with diaeresis, U+0178 ISOlat2" },
1736 * Anything below should really be kept as entities references
1738 { 402, "fnof", "latin small f with hook = function = florin, U+0192 ISOtech" },
1740 { 710, "circ", "modifier letter circumflex accent, U+02C6 ISOpub" },
1741 { 732, "tilde","small tilde, U+02DC ISOdia" },
1743 { 913, "Alpha","greek capital letter alpha, U+0391" },
1744 { 914, "Beta", "greek capital letter beta, U+0392" },
1745 { 915, "Gamma","greek capital letter gamma, U+0393 ISOgrk3" },
1746 { 916, "Delta","greek capital letter delta, U+0394 ISOgrk3" },
1747 { 917, "Epsilon","greek capital letter epsilon, U+0395" },
1748 { 918, "Zeta", "greek capital letter zeta, U+0396" },
1749 { 919, "Eta", "greek capital letter eta, U+0397" },
1750 { 920, "Theta","greek capital letter theta, U+0398 ISOgrk3" },
1751 { 921, "Iota", "greek capital letter iota, U+0399" },
1752 { 922, "Kappa","greek capital letter kappa, U+039A" },
1753 { 923, "Lambda", "greek capital letter lambda, U+039B ISOgrk3" },
1754 { 924, "Mu", "greek capital letter mu, U+039C" },
1755 { 925, "Nu", "greek capital letter nu, U+039D" },
1756 { 926, "Xi", "greek capital letter xi, U+039E ISOgrk3" },
1757 { 927, "Omicron","greek capital letter omicron, U+039F" },
1758 { 928, "Pi", "greek capital letter pi, U+03A0 ISOgrk3" },
1759 { 929, "Rho", "greek capital letter rho, U+03A1" },
1760 { 931, "Sigma","greek capital letter sigma, U+03A3 ISOgrk3" },
1761 { 932, "Tau", "greek capital letter tau, U+03A4" },
1762 { 933, "Upsilon","greek capital letter upsilon, U+03A5 ISOgrk3" },
1763 { 934, "Phi", "greek capital letter phi, U+03A6 ISOgrk3" },
1764 { 935, "Chi", "greek capital letter chi, U+03A7" },
1765 { 936, "Psi", "greek capital letter psi, U+03A8 ISOgrk3" },
1766 { 937, "Omega","greek capital letter omega, U+03A9 ISOgrk3" },
1768 { 945, "alpha","greek small letter alpha, U+03B1 ISOgrk3" },
1769 { 946, "beta", "greek small letter beta, U+03B2 ISOgrk3" },
1770 { 947, "gamma","greek small letter gamma, U+03B3 ISOgrk3" },
1771 { 948, "delta","greek small letter delta, U+03B4 ISOgrk3" },
1772 { 949, "epsilon","greek small letter epsilon, U+03B5 ISOgrk3" },
1773 { 950, "zeta", "greek small letter zeta, U+03B6 ISOgrk3" },
1774 { 951, "eta", "greek small letter eta, U+03B7 ISOgrk3" },
1775 { 952, "theta","greek small letter theta, U+03B8 ISOgrk3" },
1776 { 953, "iota", "greek small letter iota, U+03B9 ISOgrk3" },
1777 { 954, "kappa","greek small letter kappa, U+03BA ISOgrk3" },
1778 { 955, "lambda","greek small letter lambda, U+03BB ISOgrk3" },
1779 { 956, "mu", "greek small letter mu, U+03BC ISOgrk3" },
1780 { 957, "nu", "greek small letter nu, U+03BD ISOgrk3" },
1781 { 958, "xi", "greek small letter xi, U+03BE ISOgrk3" },
1782 { 959, "omicron","greek small letter omicron, U+03BF NEW" },
1783 { 960, "pi", "greek small letter pi, U+03C0 ISOgrk3" },
1784 { 961, "rho", "greek small letter rho, U+03C1 ISOgrk3" },
1785 { 962, "sigmaf","greek small letter final sigma, U+03C2 ISOgrk3" },
1786 { 963, "sigma","greek small letter sigma, U+03C3 ISOgrk3" },
1787 { 964, "tau", "greek small letter tau, U+03C4 ISOgrk3" },
1788 { 965, "upsilon","greek small letter upsilon, U+03C5 ISOgrk3" },
1789 { 966, "phi", "greek small letter phi, U+03C6 ISOgrk3" },
1790 { 967, "chi", "greek small letter chi, U+03C7 ISOgrk3" },
1791 { 968, "psi", "greek small letter psi, U+03C8 ISOgrk3" },
1792 { 969, "omega","greek small letter omega, U+03C9 ISOgrk3" },
1793 { 977, "thetasym","greek small letter theta symbol, U+03D1 NEW" },
1794 { 978, "upsih","greek upsilon with hook symbol, U+03D2 NEW" },
1795 { 982, "piv", "greek pi symbol, U+03D6 ISOgrk3" },
1797 { 8194, "ensp", "en space, U+2002 ISOpub" },
1798 { 8195, "emsp", "em space, U+2003 ISOpub" },
1799 { 8201, "thinsp","thin space, U+2009 ISOpub" },
1800 { 8204, "zwnj", "zero width non-joiner, U+200C NEW RFC 2070" },
1801 { 8205, "zwj", "zero width joiner, U+200D NEW RFC 2070" },
1802 { 8206, "lrm", "left-to-right mark, U+200E NEW RFC 2070" },
1803 { 8207, "rlm", "right-to-left mark, U+200F NEW RFC 2070" },
1804 { 8211, "ndash","en dash, U+2013 ISOpub" },
1805 { 8212, "mdash","em dash, U+2014 ISOpub" },
1806 { 8216, "lsquo","left single quotation mark, U+2018 ISOnum" },
1807 { 8217, "rsquo","right single quotation mark, U+2019 ISOnum" },
1808 { 8218, "sbquo","single low-9 quotation mark, U+201A NEW" },
1809 { 8220, "ldquo","left double quotation mark, U+201C ISOnum" },
1810 { 8221, "rdquo","right double quotation mark, U+201D ISOnum" },
1811 { 8222, "bdquo","double low-9 quotation mark, U+201E NEW" },
1812 { 8224, "dagger","dagger, U+2020 ISOpub" },
1813 { 8225, "Dagger","double dagger, U+2021 ISOpub" },
1815 { 8226, "bull", "bullet = black small circle, U+2022 ISOpub" },
1816 { 8230, "hellip","horizontal ellipsis = three dot leader, U+2026 ISOpub" },
1818 { 8240, "permil","per mille sign, U+2030 ISOtech" },
1820 { 8242, "prime","prime = minutes = feet, U+2032 ISOtech" },
1821 { 8243, "Prime","double prime = seconds = inches, U+2033 ISOtech" },
1823 { 8249, "lsaquo","single left-pointing angle quotation mark, U+2039 ISO proposed" },
1824 { 8250, "rsaquo","single right-pointing angle quotation mark, U+203A ISO proposed" },
1826 { 8254, "oline","overline = spacing overscore, U+203E NEW" },
1827 { 8260, "frasl","fraction slash, U+2044 NEW" },
1829 { 8364, "euro", "euro sign, U+20AC NEW" },
1831 { 8465, "image","blackletter capital I = imaginary part, U+2111 ISOamso" },
1832 { 8472, "weierp","script capital P = power set = Weierstrass p, U+2118 ISOamso" },
1833 { 8476, "real", "blackletter capital R = real part symbol, U+211C ISOamso" },
1834 { 8482, "trade","trade mark sign, U+2122 ISOnum" },
1835 { 8501, "alefsym","alef symbol = first transfinite cardinal, U+2135 NEW" },
1836 { 8592, "larr", "leftwards arrow, U+2190 ISOnum" },
1837 { 8593, "uarr", "upwards arrow, U+2191 ISOnum" },
1838 { 8594, "rarr", "rightwards arrow, U+2192 ISOnum" },
1839 { 8595, "darr", "downwards arrow, U+2193 ISOnum" },
1840 { 8596, "harr", "left right arrow, U+2194 ISOamsa" },
1841 { 8629, "crarr","downwards arrow with corner leftwards = carriage return, U+21B5 NEW" },
1842 { 8656, "lArr", "leftwards double arrow, U+21D0 ISOtech" },
1843 { 8657, "uArr", "upwards double arrow, U+21D1 ISOamsa" },
1844 { 8658, "rArr", "rightwards double arrow, U+21D2 ISOtech" },
1845 { 8659, "dArr", "downwards double arrow, U+21D3 ISOamsa" },
1846 { 8660, "hArr", "left right double arrow, U+21D4 ISOamsa" },
1848 { 8704, "forall","for all, U+2200 ISOtech" },
1849 { 8706, "part", "partial differential, U+2202 ISOtech" },
1850 { 8707, "exist","there exists, U+2203 ISOtech" },
1851 { 8709, "empty","empty set = null set = diameter, U+2205 ISOamso" },
1852 { 8711, "nabla","nabla = backward difference, U+2207 ISOtech" },
1853 { 8712, "isin", "element of, U+2208 ISOtech" },
1854 { 8713, "notin","not an element of, U+2209 ISOtech" },
1855 { 8715, "ni", "contains as member, U+220B ISOtech" },
1856 { 8719, "prod", "n-ary product = product sign, U+220F ISOamsb" },
1857 { 8721, "sum", "n-ary summation, U+2211 ISOamsb" },
1858 { 8722, "minus","minus sign, U+2212 ISOtech" },
1859 { 8727, "lowast","asterisk operator, U+2217 ISOtech" },
1860 { 8730, "radic","square root = radical sign, U+221A ISOtech" },
1861 { 8733, "prop", "proportional to, U+221D ISOtech" },
1862 { 8734, "infin","infinity, U+221E ISOtech" },
1863 { 8736, "ang", "angle, U+2220 ISOamso" },
1864 { 8743, "and", "logical and = wedge, U+2227 ISOtech" },
1865 { 8744, "or", "logical or = vee, U+2228 ISOtech" },
1866 { 8745, "cap", "intersection = cap, U+2229 ISOtech" },
1867 { 8746, "cup", "union = cup, U+222A ISOtech" },
1868 { 8747, "int", "integral, U+222B ISOtech" },
1869 { 8756, "there4","therefore, U+2234 ISOtech" },
1870 { 8764, "sim", "tilde operator = varies with = similar to, U+223C ISOtech" },
1871 { 8773, "cong", "approximately equal to, U+2245 ISOtech" },
1872 { 8776, "asymp","almost equal to = asymptotic to, U+2248 ISOamsr" },
1873 { 8800, "ne", "not equal to, U+2260 ISOtech" },
1874 { 8801, "equiv","identical to, U+2261 ISOtech" },
1875 { 8804, "le", "less-than or equal to, U+2264 ISOtech" },
1876 { 8805, "ge", "greater-than or equal to, U+2265 ISOtech" },
1877 { 8834, "sub", "subset of, U+2282 ISOtech" },
1878 { 8835, "sup", "superset of, U+2283 ISOtech" },
1879 { 8836, "nsub", "not a subset of, U+2284 ISOamsn" },
1880 { 8838, "sube", "subset of or equal to, U+2286 ISOtech" },
1881 { 8839, "supe", "superset of or equal to, U+2287 ISOtech" },
1882 { 8853, "oplus","circled plus = direct sum, U+2295 ISOamsb" },
1883 { 8855, "otimes","circled times = vector product, U+2297 ISOamsb" },
1884 { 8869, "perp", "up tack = orthogonal to = perpendicular, U+22A5 ISOtech" },
1885 { 8901, "sdot", "dot operator, U+22C5 ISOamsb" },
1886 { 8968, "lceil","left ceiling = apl upstile, U+2308 ISOamsc" },
1887 { 8969, "rceil","right ceiling, U+2309 ISOamsc" },
1888 { 8970, "lfloor","left floor = apl downstile, U+230A ISOamsc" },
1889 { 8971, "rfloor","right floor, U+230B ISOamsc" },
1890 { 9001, "lang", "left-pointing angle bracket = bra, U+2329 ISOtech" },
1891 { 9002, "rang", "right-pointing angle bracket = ket, U+232A ISOtech" },
1892 { 9674, "loz", "lozenge, U+25CA ISOpub" },
1894 { 9824, "spades","black spade suit, U+2660 ISOpub" },
1895 { 9827, "clubs","black club suit = shamrock, U+2663 ISOpub" },
1896 { 9829, "hearts","black heart suit = valentine, U+2665 ISOpub" },
1897 { 9830, "diams","black diamond suit, U+2666 ISOpub" },
1901 /************************************************************************
1903 * Commodity functions to handle entities *
1905 ************************************************************************/
1908 * Macro used to grow the current buffer.
1910 #define growBuffer(buffer) { \
1912 buffer##_size *= 2; \
1913 tmp = (xmlChar *) xmlRealloc(buffer, buffer##_size * sizeof(xmlChar)); \
1914 if (tmp == NULL) { \
1915 htmlErrMemory(ctxt, "growing buffer\n"); \
1924 * @name: the entity name
1926 * Lookup the given entity in EntitiesTable
1928 * TODO: the linear scan is really ugly, an hash table is really needed.
1930 * Returns the associated htmlEntityDescPtr if found, NULL otherwise.
1932 const htmlEntityDesc *
1933 htmlEntityLookup(const xmlChar *name) {
1936 for (i = 0;i < (sizeof(html40EntitiesTable)/
1937 sizeof(html40EntitiesTable[0]));i++) {
1938 if (xmlStrEqual(name, BAD_CAST html40EntitiesTable[i].name)) {
1939 return((htmlEntityDescPtr) &html40EntitiesTable[i]);
1946 * htmlEntityValueLookup:
1947 * @value: the entity's unicode value
1949 * Lookup the given entity in EntitiesTable
1951 * TODO: the linear scan is really ugly, an hash table is really needed.
1953 * Returns the associated htmlEntityDescPtr if found, NULL otherwise.
1955 const htmlEntityDesc *
1956 htmlEntityValueLookup(unsigned int value) {
1959 for (i = 0;i < (sizeof(html40EntitiesTable)/
1960 sizeof(html40EntitiesTable[0]));i++) {
1961 if (html40EntitiesTable[i].value >= value) {
1962 if (html40EntitiesTable[i].value > value)
1964 return((htmlEntityDescPtr) &html40EntitiesTable[i]);
1972 * @out: a pointer to an array of bytes to store the result
1973 * @outlen: the length of @out
1974 * @in: a pointer to an array of UTF-8 chars
1975 * @inlen: the length of @in
1977 * Take a block of UTF-8 chars in and try to convert it to an ASCII
1978 * plus HTML entities block of chars out.
1980 * Returns 0 if success, -2 if the transcoding fails, or -1 otherwise
1981 * The value of @inlen after return is the number of octets consumed
1982 * as the return value is positive, else unpredictable.
1983 * The value of @outlen after return is the number of octets consumed.
1986 UTF8ToHtml(unsigned char* out, int *outlen,
1987 const unsigned char* in, int *inlen) {
1988 const unsigned char* processed = in;
1989 const unsigned char* outend;
1990 const unsigned char* outstart = out;
1991 const unsigned char* instart = in;
1992 const unsigned char* inend;
1996 if ((out == NULL) || (outlen == NULL) || (inlen == NULL)) return(-1);
1999 * initialization nothing to do
2005 inend = in + (*inlen);
2006 outend = out + (*outlen);
2007 while (in < inend) {
2009 if (d < 0x80) { c= d; trailing= 0; }
2010 else if (d < 0xC0) {
2011 /* trailing byte in leading position */
2012 *outlen = out - outstart;
2013 *inlen = processed - instart;
2015 } else if (d < 0xE0) { c= d & 0x1F; trailing= 1; }
2016 else if (d < 0xF0) { c= d & 0x0F; trailing= 2; }
2017 else if (d < 0xF8) { c= d & 0x07; trailing= 3; }
2019 /* no chance for this in Ascii */
2020 *outlen = out - outstart;
2021 *inlen = processed - instart;
2025 if (inend - in < trailing) {
2029 for ( ; trailing; trailing--) {
2030 if ((in >= inend) || (((d= *in++) & 0xC0) != 0x80))
2036 /* assertion: c is a single UTF-4 value */
2038 if (out + 1 >= outend)
2043 const htmlEntityDesc * ent;
2048 * Try to lookup a predefined HTML entity for it
2051 ent = htmlEntityValueLookup(c);
2053 snprintf(nbuf, sizeof(nbuf), "#%u", c);
2059 if (out + 2 + len >= outend)
2062 memcpy(out, cp, len);
2068 *outlen = out - outstart;
2069 *inlen = processed - instart;
2074 * htmlEncodeEntities:
2075 * @out: a pointer to an array of bytes to store the result
2076 * @outlen: the length of @out
2077 * @in: a pointer to an array of UTF-8 chars
2078 * @inlen: the length of @in
2079 * @quoteChar: the quote character to escape (' or ") or zero.
2081 * Take a block of UTF-8 chars in and try to convert it to an ASCII
2082 * plus HTML entities block of chars out.
2084 * Returns 0 if success, -2 if the transcoding fails, or -1 otherwise
2085 * The value of @inlen after return is the number of octets consumed
2086 * as the return value is positive, else unpredictable.
2087 * The value of @outlen after return is the number of octets consumed.
2090 htmlEncodeEntities(unsigned char* out, int *outlen,
2091 const unsigned char* in, int *inlen, int quoteChar) {
2092 const unsigned char* processed = in;
2093 const unsigned char* outend;
2094 const unsigned char* outstart = out;
2095 const unsigned char* instart = in;
2096 const unsigned char* inend;
2100 if ((out == NULL) || (outlen == NULL) || (inlen == NULL) || (in == NULL))
2102 outend = out + (*outlen);
2103 inend = in + (*inlen);
2104 while (in < inend) {
2106 if (d < 0x80) { c= d; trailing= 0; }
2107 else if (d < 0xC0) {
2108 /* trailing byte in leading position */
2109 *outlen = out - outstart;
2110 *inlen = processed - instart;
2112 } else if (d < 0xE0) { c= d & 0x1F; trailing= 1; }
2113 else if (d < 0xF0) { c= d & 0x0F; trailing= 2; }
2114 else if (d < 0xF8) { c= d & 0x07; trailing= 3; }
2116 /* no chance for this in Ascii */
2117 *outlen = out - outstart;
2118 *inlen = processed - instart;
2122 if (inend - in < trailing)
2125 while (trailing--) {
2126 if (((d= *in++) & 0xC0) != 0x80) {
2127 *outlen = out - outstart;
2128 *inlen = processed - instart;
2135 /* assertion: c is a single UTF-4 value */
2136 if ((c < 0x80) && (c != (unsigned int) quoteChar) &&
2137 (c != '&') && (c != '<') && (c != '>')) {
2142 const htmlEntityDesc * ent;
2148 * Try to lookup a predefined HTML entity for it
2150 ent = htmlEntityValueLookup(c);
2152 snprintf(nbuf, sizeof(nbuf), "#%u", c);
2158 if (out + 2 + len > outend)
2161 memcpy(out, cp, len);
2167 *outlen = out - outstart;
2168 *inlen = processed - instart;
2172 /************************************************************************
2174 * Commodity functions to handle streams *
2176 ************************************************************************/
2179 * htmlNewInputStream:
2180 * @ctxt: an HTML parser context
2182 * Create a new input stream structure
2183 * Returns the new input stream or NULL
2185 static htmlParserInputPtr
2186 htmlNewInputStream(htmlParserCtxtPtr ctxt) {
2187 htmlParserInputPtr input;
2189 input = (xmlParserInputPtr) xmlMalloc(sizeof(htmlParserInput));
2190 if (input == NULL) {
2191 htmlErrMemory(ctxt, "couldn't allocate a new input stream\n");
2194 memset(input, 0, sizeof(htmlParserInput));
2195 input->filename = NULL;
2196 input->directory = NULL;
2204 input->version = NULL;
2205 input->consumed = 0;
2211 /************************************************************************
2213 * Commodity functions, cleanup needed ? *
2215 ************************************************************************/
2217 * all tags allowing pc data from the html 4.01 loose dtd
2218 * NOTE: it might be more apropriate to integrate this information
2219 * into the html40ElementTable array but I don't want to risk any
2220 * binary incomptibility
2222 static const char *allowPCData[] = {
2223 "a", "abbr", "acronym", "address", "applet", "b", "bdo", "big",
2224 "blockquote", "body", "button", "caption", "center", "cite", "code",
2225 "dd", "del", "dfn", "div", "dt", "em", "font", "form", "h1", "h2",
2226 "h3", "h4", "h5", "h6", "i", "iframe", "ins", "kbd", "label", "legend",
2227 "li", "noframes", "noscript", "object", "p", "pre", "q", "s", "samp",
2228 "small", "span", "strike", "strong", "td", "th", "tt", "u", "var"
2233 * @ctxt: an HTML parser context
2235 * @len: the size of @str
2237 * Is this a sequence of blank chars that one can ignore ?
2239 * Returns 1 if ignorable 0 otherwise.
2242 static int areBlanks(htmlParserCtxtPtr ctxt, const xmlChar *str, int len) {
2245 xmlNodePtr lastChild;
2248 for (j = 0;j < len;j++)
2249 if (!(IS_BLANK_CH(str[j]))) return(0);
2251 if (CUR == 0) return(1);
2252 if (CUR != '<') return(0);
2253 if (ctxt->name == NULL)
2255 if (xmlStrEqual(ctxt->name, BAD_CAST"html"))
2257 if (xmlStrEqual(ctxt->name, BAD_CAST"head"))
2260 /* Only strip CDATA children of the body tag for strict HTML DTDs */
2261 if (xmlStrEqual(ctxt->name, BAD_CAST "body") && ctxt->myDoc != NULL) {
2262 dtd = xmlGetIntSubset(ctxt->myDoc);
2263 if (dtd != NULL && dtd->ExternalID != NULL) {
2264 if (!xmlStrcasecmp(dtd->ExternalID, BAD_CAST "-//W3C//DTD HTML 4.01//EN") ||
2265 !xmlStrcasecmp(dtd->ExternalID, BAD_CAST "-//W3C//DTD HTML 4//EN"))
2270 if (ctxt->node == NULL) return(0);
2271 lastChild = xmlGetLastChild(ctxt->node);
2272 while ((lastChild) && (lastChild->type == XML_COMMENT_NODE))
2273 lastChild = lastChild->prev;
2274 if (lastChild == NULL) {
2275 if ((ctxt->node->type != XML_ELEMENT_NODE) &&
2276 (ctxt->node->content != NULL)) return(0);
2277 /* keep ws in constructs like ...<b> </b>...
2278 for all tags "b" allowing PCDATA */
2279 for ( i = 0; i < sizeof(allowPCData)/sizeof(allowPCData[0]); i++ ) {
2280 if ( xmlStrEqual(ctxt->name, BAD_CAST allowPCData[i]) ) {
2284 } else if (xmlNodeIsText(lastChild)) {
2287 /* keep ws in constructs like <p><b>xy</b> <i>z</i><p>
2288 for all tags "p" allowing PCDATA */
2289 for ( i = 0; i < sizeof(allowPCData)/sizeof(allowPCData[0]); i++ ) {
2290 if ( xmlStrEqual(lastChild->name, BAD_CAST allowPCData[i]) ) {
2300 * @URI: URI for the dtd, or NULL
2301 * @ExternalID: the external ID of the DTD, or NULL
2303 * Creates a new HTML document without a DTD node if @URI and @ExternalID
2306 * Returns a new document, do not initialize the DTD if not provided
2309 htmlNewDocNoDtD(const xmlChar *URI, const xmlChar *ExternalID) {
2313 * Allocate a new document and fill the fields.
2315 cur = (xmlDocPtr) xmlMalloc(sizeof(xmlDoc));
2317 htmlErrMemory(NULL, "HTML document creation failed\n");
2320 memset(cur, 0, sizeof(xmlDoc));
2322 cur->type = XML_HTML_DOCUMENT_NODE;
2323 cur->version = NULL;
2324 cur->intSubset = NULL;
2327 cur->children = NULL;
2328 cur->extSubset = NULL;
2330 cur->encoding = NULL;
2331 cur->standalone = 1;
2332 cur->compression = 0;
2335 cur->_private = NULL;
2336 cur->charset = XML_CHAR_ENCODING_UTF8;
2337 cur->properties = XML_DOC_HTML | XML_DOC_USERBUILT;
2338 if ((ExternalID != NULL) ||
2340 xmlCreateIntSubset(cur, BAD_CAST "html", ExternalID, URI);
2346 * @URI: URI for the dtd, or NULL
2347 * @ExternalID: the external ID of the DTD, or NULL
2349 * Creates a new HTML document
2351 * Returns a new document
2354 htmlNewDoc(const xmlChar *URI, const xmlChar *ExternalID) {
2355 if ((URI == NULL) && (ExternalID == NULL))
2356 return(htmlNewDocNoDtD(
2357 BAD_CAST "http://www.w3.org/TR/REC-html40/loose.dtd",
2358 BAD_CAST "-//W3C//DTD HTML 4.0 Transitional//EN"));
2360 return(htmlNewDocNoDtD(URI, ExternalID));
2364 /************************************************************************
2366 * The parser itself *
2367 * Relates to http://www.w3.org/TR/html40 *
2369 ************************************************************************/
2371 /************************************************************************
2373 * The parser itself *
2375 ************************************************************************/
2377 static const xmlChar * htmlParseNameComplex(xmlParserCtxtPtr ctxt);
2380 * htmlParseHTMLName:
2381 * @ctxt: an HTML parser context
2383 * parse an HTML tag or attribute name, note that we convert it to lowercase
2384 * since HTML names are not case-sensitive.
2386 * Returns the Tag Name parsed or NULL
2389 static const xmlChar *
2390 htmlParseHTMLName(htmlParserCtxtPtr ctxt) {
2392 xmlChar loc[HTML_PARSER_BUFFER_SIZE];
2394 if (!IS_ASCII_LETTER(CUR) && (CUR != '_') &&
2395 (CUR != ':') && (CUR != '.')) return(NULL);
2397 while ((i < HTML_PARSER_BUFFER_SIZE) &&
2398 ((IS_ASCII_LETTER(CUR)) || (IS_ASCII_DIGIT(CUR)) ||
2399 (CUR == ':') || (CUR == '-') || (CUR == '_') ||
2401 if ((CUR >= 'A') && (CUR <= 'Z')) loc[i] = CUR + 0x20;
2408 return(xmlDictLookup(ctxt->dict, loc, i));
2413 * htmlParseHTMLName_nonInvasive:
2414 * @ctxt: an HTML parser context
2416 * parse an HTML tag or attribute name, note that we convert it to lowercase
2417 * since HTML names are not case-sensitive, this doesn't consume the data
2418 * from the stream, it's a look-ahead
2420 * Returns the Tag Name parsed or NULL
2423 static const xmlChar *
2424 htmlParseHTMLName_nonInvasive(htmlParserCtxtPtr ctxt) {
2426 xmlChar loc[HTML_PARSER_BUFFER_SIZE];
2428 if (!IS_ASCII_LETTER(NXT(1)) && (NXT(1) != '_') &&
2429 (NXT(1) != ':')) return(NULL);
2431 while ((i < HTML_PARSER_BUFFER_SIZE) &&
2432 ((IS_ASCII_LETTER(NXT(1+i))) || (IS_ASCII_DIGIT(NXT(1+i))) ||
2433 (NXT(1+i) == ':') || (NXT(1+i) == '-') || (NXT(1+i) == '_'))) {
2434 if ((NXT(1+i) >= 'A') && (NXT(1+i) <= 'Z')) loc[i] = NXT(1+i) + 0x20;
2435 else loc[i] = NXT(1+i);
2439 return(xmlDictLookup(ctxt->dict, loc, i));
2445 * @ctxt: an HTML parser context
2447 * parse an HTML name, this routine is case sensitive.
2449 * Returns the Name parsed or NULL
2452 static const xmlChar *
2453 htmlParseName(htmlParserCtxtPtr ctxt) {
2461 * Accelerator for simple ASCII names
2463 in = ctxt->input->cur;
2464 if (((*in >= 0x61) && (*in <= 0x7A)) ||
2465 ((*in >= 0x41) && (*in <= 0x5A)) ||
2466 (*in == '_') || (*in == ':')) {
2468 while (((*in >= 0x61) && (*in <= 0x7A)) ||
2469 ((*in >= 0x41) && (*in <= 0x5A)) ||
2470 ((*in >= 0x30) && (*in <= 0x39)) ||
2471 (*in == '_') || (*in == '-') ||
2472 (*in == ':') || (*in == '.'))
2474 if ((*in > 0) && (*in < 0x80)) {
2475 count = in - ctxt->input->cur;
2476 ret = xmlDictLookup(ctxt->dict, ctxt->input->cur, count);
2477 ctxt->input->cur = in;
2478 ctxt->nbChars += count;
2479 ctxt->input->col += count;
2483 return(htmlParseNameComplex(ctxt));
2486 static const xmlChar *
2487 htmlParseNameComplex(xmlParserCtxtPtr ctxt) {
2493 * Handler for more complex cases
2497 if ((c == ' ') || (c == '>') || (c == '/') || /* accelerators */
2498 (!IS_LETTER(c) && (c != '_') &&
2503 while ((c != ' ') && (c != '>') && (c != '/') && /* test bigname.xml */
2504 ((IS_LETTER(c)) || (IS_DIGIT(c)) ||
2505 (c == '.') || (c == '-') ||
2506 (c == '_') || (c == ':') ||
2507 (IS_COMBINING(c)) ||
2508 (IS_EXTENDER(c)))) {
2509 if (count++ > 100) {
2517 return(xmlDictLookup(ctxt->dict, ctxt->input->cur - len, len));
2522 * htmlParseHTMLAttribute:
2523 * @ctxt: an HTML parser context
2524 * @stop: a char stop value
2526 * parse an HTML attribute value till the stop (quote), if
2527 * stop is 0 then it stops at the first space
2529 * Returns the attribute parsed or NULL
2533 htmlParseHTMLAttribute(htmlParserCtxtPtr ctxt, const xmlChar stop) {
2534 xmlChar *buffer = NULL;
2535 int buffer_size = 0;
2536 xmlChar *out = NULL;
2537 const xmlChar *name = NULL;
2538 const xmlChar *cur = NULL;
2539 const htmlEntityDesc * ent;
2542 * allocate a translation buffer.
2544 buffer_size = HTML_PARSER_BUFFER_SIZE;
2545 buffer = (xmlChar *) xmlMallocAtomic(buffer_size * sizeof(xmlChar));
2546 if (buffer == NULL) {
2547 htmlErrMemory(ctxt, "buffer allocation failed\n");
2553 * Ok loop until we reach one of the ending chars
2555 while ((CUR != 0) && (CUR != stop)) {
2556 if ((stop == 0) && (CUR == '>')) break;
2557 if ((stop == 0) && (IS_BLANK_CH(CUR))) break;
2559 if (NXT(1) == '#') {
2563 c = htmlParseCharRef(ctxt);
2565 { *out++ = c; bits= -6; }
2567 { *out++ =((c >> 6) & 0x1F) | 0xC0; bits= 0; }
2568 else if (c < 0x10000)
2569 { *out++ =((c >> 12) & 0x0F) | 0xE0; bits= 6; }
2571 { *out++ =((c >> 18) & 0x07) | 0xF0; bits= 12; }
2573 for ( ; bits >= 0; bits-= 6) {
2574 *out++ = ((c >> bits) & 0x3F) | 0x80;
2577 if (out - buffer > buffer_size - 100) {
2578 int indx = out - buffer;
2581 out = &buffer[indx];
2584 ent = htmlParseEntityRef(ctxt, &name);
2587 if (out - buffer > buffer_size - 100) {
2588 int indx = out - buffer;
2591 out = &buffer[indx];
2593 } else if (ent == NULL) {
2597 if (out - buffer > buffer_size - 100) {
2598 int indx = out - buffer;
2601 out = &buffer[indx];
2609 if (out - buffer > buffer_size - 100) {
2610 int indx = out - buffer;
2613 out = &buffer[indx];
2617 { *out++ = c; bits= -6; }
2619 { *out++ =((c >> 6) & 0x1F) | 0xC0; bits= 0; }
2620 else if (c < 0x10000)
2621 { *out++ =((c >> 12) & 0x0F) | 0xE0; bits= 6; }
2623 { *out++ =((c >> 18) & 0x07) | 0xF0; bits= 12; }
2625 for ( ; bits >= 0; bits-= 6) {
2626 *out++ = ((c >> bits) & 0x3F) | 0x80;
2634 if (out - buffer > buffer_size - 100) {
2635 int indx = out - buffer;
2638 out = &buffer[indx];
2642 { *out++ = c; bits= -6; }
2644 { *out++ =((c >> 6) & 0x1F) | 0xC0; bits= 0; }
2645 else if (c < 0x10000)
2646 { *out++ =((c >> 12) & 0x0F) | 0xE0; bits= 6; }
2648 { *out++ =((c >> 18) & 0x07) | 0xF0; bits= 12; }
2650 for ( ; bits >= 0; bits-= 6) {
2651 *out++ = ((c >> bits) & 0x3F) | 0x80;
2661 * htmlParseEntityRef:
2662 * @ctxt: an HTML parser context
2663 * @str: location to store the entity name
2665 * parse an HTML ENTITY references
2667 * [68] EntityRef ::= '&' Name ';'
2669 * Returns the associated htmlEntityDescPtr if found, or NULL otherwise,
2670 * if non-NULL *str will have to be freed by the caller.
2672 const htmlEntityDesc *
2673 htmlParseEntityRef(htmlParserCtxtPtr ctxt, const xmlChar **str) {
2674 const xmlChar *name;
2675 const htmlEntityDesc * ent = NULL;
2677 if (str != NULL) *str = NULL;
2678 if ((ctxt == NULL) || (ctxt->input == NULL)) return(NULL);
2682 name = htmlParseName(ctxt);
2684 htmlParseErr(ctxt, XML_ERR_NAME_REQUIRED,
2685 "htmlParseEntityRef: no name\n", NULL, NULL);
2693 * Lookup the entity in the table.
2695 ent = htmlEntityLookup(name);
2696 if (ent != NULL) /* OK that's ugly !!! */
2699 htmlParseErr(ctxt, XML_ERR_ENTITYREF_SEMICOL_MISSING,
2700 "htmlParseEntityRef: expecting ';'\n",
2711 * htmlParseAttValue:
2712 * @ctxt: an HTML parser context
2714 * parse a value for an attribute
2715 * Note: the parser won't do substitution of entities here, this
2716 * will be handled later in xmlStringGetNodeList, unless it was
2717 * asked for ctxt->replaceEntities != 0
2719 * Returns the AttValue parsed or NULL.
2723 htmlParseAttValue(htmlParserCtxtPtr ctxt) {
2724 xmlChar *ret = NULL;
2728 ret = htmlParseHTMLAttribute(ctxt, '"');
2730 htmlParseErr(ctxt, XML_ERR_ATTRIBUTE_NOT_FINISHED,
2731 "AttValue: \" expected\n", NULL, NULL);
2734 } else if (CUR == '\'') {
2736 ret = htmlParseHTMLAttribute(ctxt, '\'');
2738 htmlParseErr(ctxt, XML_ERR_ATTRIBUTE_NOT_FINISHED,
2739 "AttValue: ' expected\n", NULL, NULL);
2744 * That's an HTMLism, the attribute value may not be quoted
2746 ret = htmlParseHTMLAttribute(ctxt, 0);
2748 htmlParseErr(ctxt, XML_ERR_ATTRIBUTE_WITHOUT_VALUE,
2749 "AttValue: no value found\n", NULL, NULL);
2756 * htmlParseSystemLiteral:
2757 * @ctxt: an HTML parser context
2759 * parse an HTML Literal
2761 * [11] SystemLiteral ::= ('"' [^"]* '"') | ("'" [^']* "'")
2763 * Returns the SystemLiteral parsed or NULL
2767 htmlParseSystemLiteral(htmlParserCtxtPtr ctxt) {
2769 xmlChar *ret = NULL;
2774 while ((IS_CHAR_CH(CUR)) && (CUR != '"'))
2776 if (!IS_CHAR_CH(CUR)) {
2777 htmlParseErr(ctxt, XML_ERR_LITERAL_NOT_FINISHED,
2778 "Unfinished SystemLiteral\n", NULL, NULL);
2780 ret = xmlStrndup(q, CUR_PTR - q);
2783 } else if (CUR == '\'') {
2786 while ((IS_CHAR_CH(CUR)) && (CUR != '\''))
2788 if (!IS_CHAR_CH(CUR)) {
2789 htmlParseErr(ctxt, XML_ERR_LITERAL_NOT_FINISHED,
2790 "Unfinished SystemLiteral\n", NULL, NULL);
2792 ret = xmlStrndup(q, CUR_PTR - q);
2796 htmlParseErr(ctxt, XML_ERR_LITERAL_NOT_STARTED,
2797 " or ' expected\n", NULL, NULL);
2804 * htmlParsePubidLiteral:
2805 * @ctxt: an HTML parser context
2807 * parse an HTML public literal
2809 * [12] PubidLiteral ::= '"' PubidChar* '"' | "'" (PubidChar - "'")* "'"
2811 * Returns the PubidLiteral parsed or NULL.
2815 htmlParsePubidLiteral(htmlParserCtxtPtr ctxt) {
2817 xmlChar *ret = NULL;
2819 * Name ::= (Letter | '_') (NameChar)*
2824 while (IS_PUBIDCHAR_CH(CUR)) NEXT;
2826 htmlParseErr(ctxt, XML_ERR_LITERAL_NOT_FINISHED,
2827 "Unfinished PubidLiteral\n", NULL, NULL);
2829 ret = xmlStrndup(q, CUR_PTR - q);
2832 } else if (CUR == '\'') {
2835 while ((IS_PUBIDCHAR_CH(CUR)) && (CUR != '\''))
2838 htmlParseErr(ctxt, XML_ERR_LITERAL_NOT_FINISHED,
2839 "Unfinished PubidLiteral\n", NULL, NULL);
2841 ret = xmlStrndup(q, CUR_PTR - q);
2845 htmlParseErr(ctxt, XML_ERR_LITERAL_NOT_STARTED,
2846 "PubidLiteral \" or ' expected\n", NULL, NULL);
2854 * @ctxt: an HTML parser context
2856 * parse the content of an HTML SCRIPT or STYLE element
2857 * http://www.w3.org/TR/html4/sgml/dtd.html#Script
2858 * http://www.w3.org/TR/html4/sgml/dtd.html#StyleSheet
2859 * http://www.w3.org/TR/html4/types.html#type-script
2860 * http://www.w3.org/TR/html4/types.html#h-6.15
2861 * http://www.w3.org/TR/html4/appendix/notes.html#h-B.3.2.1
2863 * Script data ( %Script; in the DTD) can be the content of the SCRIPT
2864 * element and the value of intrinsic event attributes. User agents must
2865 * not evaluate script data as HTML markup but instead must pass it on as
2866 * data to a script engine.
2868 * - The content is passed like CDATA
2869 * - the attributes for style and scripting "onXXX" are also described
2870 * as CDATA but SGML allows entities references in attributes so their
2871 * processing is identical as other attributes
2874 htmlParseScript(htmlParserCtxtPtr ctxt) {
2875 xmlChar buf[HTML_PARSER_BIG_BUFFER_SIZE + 5];
2881 while (IS_CHAR_CH(cur)) {
2882 if ((cur == '<') && (NXT(1) == '/')) {
2884 * One should break here, the specification is clear:
2885 * Authors should therefore escape "</" within the content.
2886 * Escape mechanisms are specific to each scripting or
2887 * style sheet language.
2889 * In recovery mode, only break if end tag match the
2890 * current tag, effectively ignoring all tags inside the
2891 * script/style block and treating the entire block as
2894 if (ctxt->recovery) {
2895 if (xmlStrncasecmp(ctxt->name, ctxt->input->cur+2,
2896 xmlStrlen(ctxt->name)) == 0)
2900 htmlParseErr(ctxt, XML_ERR_TAG_NAME_MISMATCH,
2901 "Element %s embeds close tag\n",
2905 if (((NXT(2) >= 'A') && (NXT(2) <= 'Z')) ||
2906 ((NXT(2) >= 'a') && (NXT(2) <= 'z')))
2912 COPY_BUF(l,buf,nbchar,cur);
2913 if (nbchar >= HTML_PARSER_BIG_BUFFER_SIZE) {
2914 if (ctxt->sax->cdataBlock!= NULL) {
2916 * Insert as CDATA, which is the same as HTML_PRESERVE_NODE
2918 ctxt->sax->cdataBlock(ctxt->userData, buf, nbchar);
2919 } else if (ctxt->sax->characters != NULL) {
2920 ctxt->sax->characters(ctxt->userData, buf, nbchar);
2929 if ((!(IS_CHAR_CH(cur))) && (!((cur == 0) && (ctxt->progressive)))) {
2930 htmlParseErrInt(ctxt, XML_ERR_INVALID_CHAR,
2931 "Invalid char in CDATA 0x%X\n", cur);
2932 if (ctxt->input->cur < ctxt->input->end) {
2937 if ((nbchar != 0) && (ctxt->sax != NULL) && (!ctxt->disableSAX)) {
2938 if (ctxt->sax->cdataBlock!= NULL) {
2940 * Insert as CDATA, which is the same as HTML_PRESERVE_NODE
2942 ctxt->sax->cdataBlock(ctxt->userData, buf, nbchar);
2943 } else if (ctxt->sax->characters != NULL) {
2944 ctxt->sax->characters(ctxt->userData, buf, nbchar);
2951 * htmlParseCharDataInternal:
2952 * @ctxt: an HTML parser context
2953 * @readahead: optional read ahead character in ascii range
2955 * parse a CharData section.
2956 * if we are within a CDATA section ']]>' marks an end of section.
2958 * [14] CharData ::= [^<&]* - ([^<&]* ']]>' [^<&]*)
2962 htmlParseCharDataInternal(htmlParserCtxtPtr ctxt, int readahead) {
2963 xmlChar buf[HTML_PARSER_BIG_BUFFER_SIZE + 6];
2969 buf[nbchar++] = readahead;
2973 while (((cur != '<') || (ctxt->token == '<')) &&
2974 ((cur != '&') || (ctxt->token == '&')) &&
2976 if (!(IS_CHAR(cur))) {
2977 htmlParseErrInt(ctxt, XML_ERR_INVALID_CHAR,
2978 "Invalid char in CDATA 0x%X\n", cur);
2980 COPY_BUF(l,buf,nbchar,cur);
2982 if (nbchar >= HTML_PARSER_BIG_BUFFER_SIZE) {
2984 * Ok the segment is to be consumed as chars.
2986 if ((ctxt->sax != NULL) && (!ctxt->disableSAX)) {
2987 if (areBlanks(ctxt, buf, nbchar)) {
2988 if (ctxt->keepBlanks) {
2989 if (ctxt->sax->characters != NULL)
2990 ctxt->sax->characters(ctxt->userData, buf, nbchar);
2992 if (ctxt->sax->ignorableWhitespace != NULL)
2993 ctxt->sax->ignorableWhitespace(ctxt->userData,
2997 htmlCheckParagraph(ctxt);
2998 if (ctxt->sax->characters != NULL)
2999 ctxt->sax->characters(ctxt->userData, buf, nbchar);
3006 if (chunk > HTML_PARSER_BUFFER_SIZE) {
3022 * Ok the segment is to be consumed as chars.
3024 if ((ctxt->sax != NULL) && (!ctxt->disableSAX)) {
3025 if (areBlanks(ctxt, buf, nbchar)) {
3026 if (ctxt->keepBlanks) {
3027 if (ctxt->sax->characters != NULL)
3028 ctxt->sax->characters(ctxt->userData, buf, nbchar);
3030 if (ctxt->sax->ignorableWhitespace != NULL)
3031 ctxt->sax->ignorableWhitespace(ctxt->userData,
3035 htmlCheckParagraph(ctxt);
3036 if (ctxt->sax->characters != NULL)
3037 ctxt->sax->characters(ctxt->userData, buf, nbchar);
3045 ctxt->instate = XML_PARSER_EOF;
3050 * htmlParseCharData:
3051 * @ctxt: an HTML parser context
3053 * parse a CharData section.
3054 * if we are within a CDATA section ']]>' marks an end of section.
3056 * [14] CharData ::= [^<&]* - ([^<&]* ']]>' [^<&]*)
3060 htmlParseCharData(htmlParserCtxtPtr ctxt) {
3061 htmlParseCharDataInternal(ctxt, 0);
3065 * htmlParseExternalID:
3066 * @ctxt: an HTML parser context
3067 * @publicID: a xmlChar** receiving PubidLiteral
3069 * Parse an External ID or a Public ID
3071 * [75] ExternalID ::= 'SYSTEM' S SystemLiteral
3072 * | 'PUBLIC' S PubidLiteral S SystemLiteral
3074 * [83] PublicID ::= 'PUBLIC' S PubidLiteral
3076 * Returns the function returns SystemLiteral and in the second
3077 * case publicID receives PubidLiteral, is strict is off
3078 * it is possible to return NULL and have publicID set.
3082 htmlParseExternalID(htmlParserCtxtPtr ctxt, xmlChar **publicID) {
3083 xmlChar *URI = NULL;
3085 if ((UPPER == 'S') && (UPP(1) == 'Y') &&
3086 (UPP(2) == 'S') && (UPP(3) == 'T') &&
3087 (UPP(4) == 'E') && (UPP(5) == 'M')) {
3089 if (!IS_BLANK_CH(CUR)) {
3090 htmlParseErr(ctxt, XML_ERR_SPACE_REQUIRED,
3091 "Space required after 'SYSTEM'\n", NULL, NULL);
3094 URI = htmlParseSystemLiteral(ctxt);
3096 htmlParseErr(ctxt, XML_ERR_URI_REQUIRED,
3097 "htmlParseExternalID: SYSTEM, no URI\n", NULL, NULL);
3099 } else if ((UPPER == 'P') && (UPP(1) == 'U') &&
3100 (UPP(2) == 'B') && (UPP(3) == 'L') &&
3101 (UPP(4) == 'I') && (UPP(5) == 'C')) {
3103 if (!IS_BLANK_CH(CUR)) {
3104 htmlParseErr(ctxt, XML_ERR_SPACE_REQUIRED,
3105 "Space required after 'PUBLIC'\n", NULL, NULL);
3108 *publicID = htmlParsePubidLiteral(ctxt);
3109 if (*publicID == NULL) {
3110 htmlParseErr(ctxt, XML_ERR_PUBID_REQUIRED,
3111 "htmlParseExternalID: PUBLIC, no Public Identifier\n",
3115 if ((CUR == '"') || (CUR == '\'')) {
3116 URI = htmlParseSystemLiteral(ctxt);
3124 * @ctxt: an XML parser context
3126 * parse an XML Processing Instruction.
3128 * [16] PI ::= '<?' PITarget (S (Char* - (Char* '?>' Char*)))? '?>'
3131 htmlParsePI(htmlParserCtxtPtr ctxt) {
3132 xmlChar *buf = NULL;
3134 int size = HTML_PARSER_BUFFER_SIZE;
3136 const xmlChar *target;
3137 xmlParserInputState state;
3140 if ((RAW == '<') && (NXT(1) == '?')) {
3141 state = ctxt->instate;
3142 ctxt->instate = XML_PARSER_PI;
3144 * this is a Processing Instruction.
3150 * Parse the target name and check for special support like
3153 target = htmlParseName(ctxt);
3154 if (target != NULL) {
3161 if ((ctxt->sax) && (!ctxt->disableSAX) &&
3162 (ctxt->sax->processingInstruction != NULL))
3163 ctxt->sax->processingInstruction(ctxt->userData,
3165 ctxt->instate = state;
3168 buf = (xmlChar *) xmlMallocAtomic(size * sizeof(xmlChar));
3170 htmlErrMemory(ctxt, NULL);
3171 ctxt->instate = state;
3175 if (!IS_BLANK(cur)) {
3176 htmlParseErr(ctxt, XML_ERR_SPACE_REQUIRED,
3177 "ParsePI: PI %s space expected\n", target, NULL);
3181 while (IS_CHAR(cur) && (cur != '>')) {
3182 if (len + 5 >= size) {
3186 tmp = (xmlChar *) xmlRealloc(buf, size * sizeof(xmlChar));
3188 htmlErrMemory(ctxt, NULL);
3190 ctxt->instate = state;
3200 COPY_BUF(l,buf,len,cur);
3211 htmlParseErr(ctxt, XML_ERR_PI_NOT_FINISHED,
3212 "ParsePI: PI %s never end ...\n", target, NULL);
3219 if ((ctxt->sax) && (!ctxt->disableSAX) &&
3220 (ctxt->sax->processingInstruction != NULL))
3221 ctxt->sax->processingInstruction(ctxt->userData,
3226 htmlParseErr(ctxt, XML_ERR_PI_NOT_STARTED,
3227 "PI is not started correctly", NULL, NULL);
3229 ctxt->instate = state;
3235 * @ctxt: an HTML parser context
3237 * Parse an XML (SGML) comment <!-- .... -->
3239 * [15] Comment ::= '<!--' ((Char - '-') | ('-' (Char - '-')))* '-->'
3242 htmlParseComment(htmlParserCtxtPtr ctxt) {
3243 xmlChar *buf = NULL;
3245 int size = HTML_PARSER_BUFFER_SIZE;
3249 xmlParserInputState state;
3252 * Check that there is a comment right here.
3254 if ((RAW != '<') || (NXT(1) != '!') ||
3255 (NXT(2) != '-') || (NXT(3) != '-')) return;
3257 state = ctxt->instate;
3258 ctxt->instate = XML_PARSER_COMMENT;
3261 buf = (xmlChar *) xmlMallocAtomic(size * sizeof(xmlChar));
3263 htmlErrMemory(ctxt, "buffer allocation failed\n");
3264 ctxt->instate = state;
3278 while (IS_CHAR(cur) &&
3280 (r != '-') || (q != '-'))) {
3281 if (len + 5 >= size) {
3285 tmp = (xmlChar *) xmlRealloc(buf, size * sizeof(xmlChar));
3288 htmlErrMemory(ctxt, "growing buffer failed\n");
3289 ctxt->instate = state;
3294 COPY_BUF(ql,buf,len,q);
3310 if ((ctxt->sax != NULL) && (ctxt->sax->comment != NULL) &&
3311 (!ctxt->disableSAX))
3312 ctxt->sax->comment(ctxt->userData, buf);
3314 ctxt->instate = state;
3319 htmlParseErr(ctxt, XML_ERR_COMMENT_NOT_FINISHED,
3320 "Comment not terminated \n<!--%.50s\n", buf, NULL);
3326 * @ctxt: an HTML parser context
3328 * parse Reference declarations
3330 * [66] CharRef ::= '&#' [0-9]+ ';' |
3331 * '&#x' [0-9a-fA-F]+ ';'
3333 * Returns the value parsed (as an int)
3336 htmlParseCharRef(htmlParserCtxtPtr ctxt) {
3339 if ((ctxt == NULL) || (ctxt->input == NULL)) {
3340 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
3341 "htmlParseCharRef: context error\n",
3345 if ((CUR == '&') && (NXT(1) == '#') &&
3346 ((NXT(2) == 'x') || NXT(2) == 'X')) {
3348 while (CUR != ';') {
3349 if ((CUR >= '0') && (CUR <= '9'))
3350 val = val * 16 + (CUR - '0');
3351 else if ((CUR >= 'a') && (CUR <= 'f'))
3352 val = val * 16 + (CUR - 'a') + 10;
3353 else if ((CUR >= 'A') && (CUR <= 'F'))
3354 val = val * 16 + (CUR - 'A') + 10;
3356 htmlParseErr(ctxt, XML_ERR_INVALID_HEX_CHARREF,
3357 "htmlParseCharRef: missing semicolon\n",
3365 } else if ((CUR == '&') && (NXT(1) == '#')) {
3367 while (CUR != ';') {
3368 if ((CUR >= '0') && (CUR <= '9'))
3369 val = val * 10 + (CUR - '0');
3371 htmlParseErr(ctxt, XML_ERR_INVALID_DEC_CHARREF,
3372 "htmlParseCharRef: missing semicolon\n",
3381 htmlParseErr(ctxt, XML_ERR_INVALID_CHARREF,
3382 "htmlParseCharRef: invalid value\n", NULL, NULL);
3385 * Check the value IS_CHAR ...
3390 htmlParseErrInt(ctxt, XML_ERR_INVALID_CHAR,
3391 "htmlParseCharRef: invalid xmlChar value %d\n",
3399 * htmlParseDocTypeDecl:
3400 * @ctxt: an HTML parser context
3402 * parse a DOCTYPE declaration
3404 * [28] doctypedecl ::= '<!DOCTYPE' S Name (S ExternalID)? S?
3405 * ('[' (markupdecl | PEReference | S)* ']' S?)? '>'
3409 htmlParseDocTypeDecl(htmlParserCtxtPtr ctxt) {
3410 const xmlChar *name;
3411 xmlChar *ExternalID = NULL;
3412 xmlChar *URI = NULL;
3415 * We know that '<!DOCTYPE' has been detected.
3422 * Parse the DOCTYPE name.
3424 name = htmlParseName(ctxt);
3426 htmlParseErr(ctxt, XML_ERR_NAME_REQUIRED,
3427 "htmlParseDocTypeDecl : no DOCTYPE name !\n",
3431 * Check that upper(name) == "HTML" !!!!!!!!!!!!!
3437 * Check for SystemID and ExternalID
3439 URI = htmlParseExternalID(ctxt, &ExternalID);
3443 * We should be at the end of the DOCTYPE declaration.
3446 htmlParseErr(ctxt, XML_ERR_DOCTYPE_NOT_FINISHED,
3447 "DOCTYPE improperly terminated\n", NULL, NULL);
3448 /* We shouldn't try to resynchronize ... */
3453 * Create or update the document accordingly to the DOCTYPE
3455 if ((ctxt->sax != NULL) && (ctxt->sax->internalSubset != NULL) &&
3456 (!ctxt->disableSAX))
3457 ctxt->sax->internalSubset(ctxt->userData, name, ExternalID, URI);
3460 * Cleanup, since we don't use all those identifiers
3462 if (URI != NULL) xmlFree(URI);
3463 if (ExternalID != NULL) xmlFree(ExternalID);
3467 * htmlParseAttribute:
3468 * @ctxt: an HTML parser context
3469 * @value: a xmlChar ** used to store the value of the attribute
3471 * parse an attribute
3473 * [41] Attribute ::= Name Eq AttValue
3475 * [25] Eq ::= S? '=' S?
3479 * [NS 11] Attribute ::= QName Eq AttValue
3481 * Also the case QName == xmlns:??? is handled independently as a namespace
3484 * Returns the attribute name, and the value in *value.
3487 static const xmlChar *
3488 htmlParseAttribute(htmlParserCtxtPtr ctxt, xmlChar **value) {
3489 const xmlChar *name;
3490 xmlChar *val = NULL;
3493 name = htmlParseHTMLName(ctxt);
3495 htmlParseErr(ctxt, XML_ERR_NAME_REQUIRED,
3496 "error parsing attribute name\n", NULL, NULL);
3507 val = htmlParseAttValue(ctxt);
3515 * htmlCheckEncodingDirect:
3516 * @ctxt: an HTML parser context
3517 * @attvalue: the attribute value
3519 * Checks an attribute value to detect
3521 * If a new encoding is detected the parser is switched to decode
3525 htmlCheckEncodingDirect(htmlParserCtxtPtr ctxt, const xmlChar *encoding) {
3527 if ((ctxt == NULL) || (encoding == NULL) ||
3528 (ctxt->options & HTML_PARSE_IGNORE_ENC))
3531 /* do not change encoding */
3532 if (ctxt->input->encoding != NULL)
3535 if (encoding != NULL) {
3536 xmlCharEncoding enc;
3537 xmlCharEncodingHandlerPtr handler;
3539 while ((*encoding == ' ') || (*encoding == '\t')) encoding++;
3541 if (ctxt->input->encoding != NULL)
3542 xmlFree((xmlChar *) ctxt->input->encoding);
3543 ctxt->input->encoding = xmlStrdup(encoding);
3545 enc = xmlParseCharEncoding((const char *) encoding);
3547 * registered set of known encodings
3549 if (enc != XML_CHAR_ENCODING_ERROR) {
3550 if (((enc == XML_CHAR_ENCODING_UTF16LE) ||
3551 (enc == XML_CHAR_ENCODING_UTF16BE) ||
3552 (enc == XML_CHAR_ENCODING_UCS4LE) ||
3553 (enc == XML_CHAR_ENCODING_UCS4BE)) &&
3554 (ctxt->input->buf != NULL) &&
3555 (ctxt->input->buf->encoder == NULL)) {
3556 htmlParseErr(ctxt, XML_ERR_INVALID_ENCODING,
3557 "htmlCheckEncoding: wrong encoding meta\n",
3560 xmlSwitchEncoding(ctxt, enc);
3562 ctxt->charset = XML_CHAR_ENCODING_UTF8;
3565 * fallback for unknown encodings
3567 handler = xmlFindCharEncodingHandler((const char *) encoding);
3568 if (handler != NULL) {
3569 xmlSwitchToEncoding(ctxt, handler);
3570 ctxt->charset = XML_CHAR_ENCODING_UTF8;
3572 htmlParseErr(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
3573 "htmlCheckEncoding: unknown encoding %s\n",
3578 if ((ctxt->input->buf != NULL) &&
3579 (ctxt->input->buf->encoder != NULL) &&
3580 (ctxt->input->buf->raw != NULL) &&
3581 (ctxt->input->buf->buffer != NULL)) {
3586 * convert as much as possible to the parser reading buffer.
3588 processed = ctxt->input->cur - ctxt->input->base;
3589 xmlBufShrink(ctxt->input->buf->buffer, processed);
3590 nbchars = xmlCharEncInput(ctxt->input->buf, 1);
3592 htmlParseErr(ctxt, XML_ERR_INVALID_ENCODING,
3593 "htmlCheckEncoding: encoder error\n",
3596 xmlBufResetInput(ctxt->input->buf->buffer, ctxt->input);
3602 * htmlCheckEncoding:
3603 * @ctxt: an HTML parser context
3604 * @attvalue: the attribute value
3606 * Checks an http-equiv attribute from a Meta tag to detect
3608 * If a new encoding is detected the parser is switched to decode
3612 htmlCheckEncoding(htmlParserCtxtPtr ctxt, const xmlChar *attvalue) {
3613 const xmlChar *encoding;
3618 encoding = xmlStrcasestr(attvalue, BAD_CAST"charset");
3619 if (encoding != NULL) {
3625 if (encoding && IS_BLANK_CH(*encoding))
3626 encoding = xmlStrcasestr(attvalue, BAD_CAST"=");
3627 if (encoding && *encoding == '=') {
3629 htmlCheckEncodingDirect(ctxt, encoding);
3635 * @ctxt: an HTML parser context
3636 * @atts: the attributes values
3638 * Checks an attributes from a Meta tag
3641 htmlCheckMeta(htmlParserCtxtPtr ctxt, const xmlChar **atts) {
3643 const xmlChar *att, *value;
3645 const xmlChar *content = NULL;
3647 if ((ctxt == NULL) || (atts == NULL))
3652 while (att != NULL) {
3654 if ((value != NULL) && (!xmlStrcasecmp(att, BAD_CAST"http-equiv"))
3655 && (!xmlStrcasecmp(value, BAD_CAST"Content-Type")))
3657 else if ((value != NULL) && (!xmlStrcasecmp(att, BAD_CAST"charset")))
3658 htmlCheckEncodingDirect(ctxt, value);
3659 else if ((value != NULL) && (!xmlStrcasecmp(att, BAD_CAST"content")))
3663 if ((http) && (content != NULL))
3664 htmlCheckEncoding(ctxt, content);
3669 * htmlParseStartTag:
3670 * @ctxt: an HTML parser context
3672 * parse a start of tag either for rule element or
3673 * EmptyElement. In both case we don't parse the tag closing chars.
3675 * [40] STag ::= '<' Name (S Attribute)* S? '>'
3677 * [44] EmptyElemTag ::= '<' Name (S Attribute)* S? '/>'
3681 * [NS 8] STag ::= '<' QName (S Attribute)* S? '>'
3683 * [NS 10] EmptyElement ::= '<' QName (S Attribute)* S? '/>'
3685 * Returns 0 in case of success, -1 in case of error and 1 if discarded
3689 htmlParseStartTag(htmlParserCtxtPtr ctxt) {
3690 const xmlChar *name;
3691 const xmlChar *attname;
3693 const xmlChar **atts;
3700 if ((ctxt == NULL) || (ctxt->input == NULL)) {
3701 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
3702 "htmlParseStartTag: context error\n", NULL, NULL);
3705 if (ctxt->instate == XML_PARSER_EOF)
3707 if (CUR != '<') return -1;
3711 maxatts = ctxt->maxatts;
3714 name = htmlParseHTMLName(ctxt);
3716 htmlParseErr(ctxt, XML_ERR_NAME_REQUIRED,
3717 "htmlParseStartTag: invalid element name\n",
3719 /* if recover preserve text on classic misconstructs */
3720 if ((ctxt->recovery) && ((IS_BLANK_CH(CUR)) || (CUR == '<') ||
3721 (CUR == '=') || (CUR == '>') || (((CUR >= '0') && (CUR <= '9'))))) {
3722 htmlParseCharDataInternal(ctxt, '<');
3727 /* Dump the bogus tag like browsers do */
3728 while ((IS_CHAR_CH(CUR)) && (CUR != '>') &&
3729 (ctxt->instate != XML_PARSER_EOF))
3733 if (xmlStrEqual(name, BAD_CAST"meta"))
3737 * Check for auto-closure of HTML elements.
3739 htmlAutoClose(ctxt, name);
3742 * Check for implied HTML elements.
3744 htmlCheckImplied(ctxt, name);
3747 * Avoid html at any level > 0, head at any level != 1
3748 * or any attempt to recurse body
3750 if ((ctxt->nameNr > 0) && (xmlStrEqual(name, BAD_CAST"html"))) {
3751 htmlParseErr(ctxt, XML_HTML_STRUCURE_ERROR,
3752 "htmlParseStartTag: misplaced <html> tag\n",
3757 if ((ctxt->nameNr != 1) &&
3758 (xmlStrEqual(name, BAD_CAST"head"))) {
3759 htmlParseErr(ctxt, XML_HTML_STRUCURE_ERROR,
3760 "htmlParseStartTag: misplaced <head> tag\n",
3765 if (xmlStrEqual(name, BAD_CAST"body")) {
3767 for (indx = 0;indx < ctxt->nameNr;indx++) {
3768 if (xmlStrEqual(ctxt->nameTab[indx], BAD_CAST"body")) {
3769 htmlParseErr(ctxt, XML_HTML_STRUCURE_ERROR,
3770 "htmlParseStartTag: misplaced <body> tag\n",
3779 * Now parse the attributes, it ends up with the ending
3784 while ((IS_CHAR_CH(CUR)) &&
3786 ((CUR != '/') || (NXT(1) != '>'))) {
3787 long cons = ctxt->nbChars;
3790 attname = htmlParseAttribute(ctxt, &attvalue);
3791 if (attname != NULL) {
3794 * Well formedness requires at most one declaration of an attribute
3796 for (i = 0; i < nbatts;i += 2) {
3797 if (xmlStrEqual(atts[i], attname)) {
3798 htmlParseErr(ctxt, XML_ERR_ATTRIBUTE_REDEFINED,
3799 "Attribute %s redefined\n", attname, NULL);
3800 if (attvalue != NULL)
3807 * Add the pair to atts
3810 maxatts = 22; /* allow for 10 attrs by default */
3811 atts = (const xmlChar **)
3812 xmlMalloc(maxatts * sizeof(xmlChar *));
3814 htmlErrMemory(ctxt, NULL);
3815 if (attvalue != NULL)
3820 ctxt->maxatts = maxatts;
3821 } else if (nbatts + 4 > maxatts) {
3825 n = (const xmlChar **) xmlRealloc((void *) atts,
3826 maxatts * sizeof(const xmlChar *));
3828 htmlErrMemory(ctxt, NULL);
3829 if (attvalue != NULL)
3835 ctxt->maxatts = maxatts;
3837 atts[nbatts++] = attname;
3838 atts[nbatts++] = attvalue;
3839 atts[nbatts] = NULL;
3840 atts[nbatts + 1] = NULL;
3843 if (attvalue != NULL)
3845 /* Dump the bogus attribute string up to the next blank or
3846 * the end of the tag. */
3847 while ((IS_CHAR_CH(CUR)) &&
3848 !(IS_BLANK_CH(CUR)) && (CUR != '>') &&
3849 ((CUR != '/') || (NXT(1) != '>')))
3855 if (cons == ctxt->nbChars) {
3856 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
3857 "htmlParseStartTag: problem parsing attributes\n",
3864 * Handle specific association to the META tag
3866 if (meta && (nbatts != 0))
3867 htmlCheckMeta(ctxt, atts);
3870 * SAX: Start of Element !
3873 htmlnamePush(ctxt, name);
3874 if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL)) {
3876 ctxt->sax->startElement(ctxt->userData, name, atts);
3878 ctxt->sax->startElement(ctxt->userData, name, NULL);
3883 for (i = 1;i < nbatts;i += 2) {
3884 if (atts[i] != NULL)
3885 xmlFree((xmlChar *) atts[i]);
3894 * @ctxt: an HTML parser context
3896 * parse an end of tag
3898 * [42] ETag ::= '</' Name S? '>'
3902 * [NS 9] ETag ::= '</' QName S? '>'
3904 * Returns 1 if the current level should be closed.
3908 htmlParseEndTag(htmlParserCtxtPtr ctxt)
3910 const xmlChar *name;
3911 const xmlChar *oldname;
3914 if ((CUR != '<') || (NXT(1) != '/')) {
3915 htmlParseErr(ctxt, XML_ERR_LTSLASH_REQUIRED,
3916 "htmlParseEndTag: '</' not found\n", NULL, NULL);
3921 name = htmlParseHTMLName(ctxt);
3925 * We should definitely be at the ending "S? '>'" part
3928 if ((!IS_CHAR_CH(CUR)) || (CUR != '>')) {
3929 htmlParseErr(ctxt, XML_ERR_GT_REQUIRED,
3930 "End tag : expected '>'\n", NULL, NULL);
3931 if (ctxt->recovery) {
3933 * We're not at the ending > !!
3934 * Error, unless in recover mode where we search forwards
3937 while (CUR != '\0' && CUR != '>') NEXT;
3944 * if we ignored misplaced tags in htmlParseStartTag don't pop them
3947 if ((ctxt->depth > 0) &&
3948 (xmlStrEqual(name, BAD_CAST "html") ||
3949 xmlStrEqual(name, BAD_CAST "body") ||
3950 xmlStrEqual(name, BAD_CAST "head"))) {
3956 * If the name read is not one of the element in the parsing stack
3957 * then return, it's just an error.
3959 for (i = (ctxt->nameNr - 1); i >= 0; i--) {
3960 if (xmlStrEqual(name, ctxt->nameTab[i]))
3964 htmlParseErr(ctxt, XML_ERR_TAG_NAME_MISMATCH,
3965 "Unexpected end tag : %s\n", name, NULL);
3971 * Check for auto-closure of HTML elements.
3974 htmlAutoCloseOnClose(ctxt, name);
3977 * Well formedness constraints, opening and closing must match.
3978 * With the exception that the autoclose may have popped stuff out
3981 if (!xmlStrEqual(name, ctxt->name)) {
3982 if ((ctxt->name != NULL) && (!xmlStrEqual(ctxt->name, name))) {
3983 htmlParseErr(ctxt, XML_ERR_TAG_NAME_MISMATCH,
3984 "Opening and ending tag mismatch: %s and %s\n",
3992 oldname = ctxt->name;
3993 if ((oldname != NULL) && (xmlStrEqual(oldname, name))) {
3994 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
3995 ctxt->sax->endElement(ctxt->userData, name);
3996 htmlNodeInfoPop(ctxt);
4008 * htmlParseReference:
4009 * @ctxt: an HTML parser context
4011 * parse and handle entity references in content,
4012 * this will end-up in a call to character() since this is either a
4013 * CharRef, or a predefined entity.
4016 htmlParseReference(htmlParserCtxtPtr ctxt) {
4017 const htmlEntityDesc * ent;
4019 const xmlChar *name;
4020 if (CUR != '&') return;
4022 if (NXT(1) == '#') {
4026 c = htmlParseCharRef(ctxt);
4030 if (c < 0x80) { out[i++]= c; bits= -6; }
4031 else if (c < 0x800) { out[i++]=((c >> 6) & 0x1F) | 0xC0; bits= 0; }
4032 else if (c < 0x10000) { out[i++]=((c >> 12) & 0x0F) | 0xE0; bits= 6; }
4033 else { out[i++]=((c >> 18) & 0x07) | 0xF0; bits= 12; }
4035 for ( ; bits >= 0; bits-= 6) {
4036 out[i++]= ((c >> bits) & 0x3F) | 0x80;
4040 htmlCheckParagraph(ctxt);
4041 if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
4042 ctxt->sax->characters(ctxt->userData, out, i);
4044 ent = htmlParseEntityRef(ctxt, &name);
4046 htmlCheckParagraph(ctxt);
4047 if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
4048 ctxt->sax->characters(ctxt->userData, BAD_CAST "&", 1);
4051 if ((ent == NULL) || !(ent->value > 0)) {
4052 htmlCheckParagraph(ctxt);
4053 if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL)) {
4054 ctxt->sax->characters(ctxt->userData, BAD_CAST "&", 1);
4055 ctxt->sax->characters(ctxt->userData, name, xmlStrlen(name));
4056 /* ctxt->sax->characters(ctxt->userData, BAD_CAST ";", 1); */
4064 { out[i++]= c; bits= -6; }
4066 { out[i++]=((c >> 6) & 0x1F) | 0xC0; bits= 0; }
4067 else if (c < 0x10000)
4068 { out[i++]=((c >> 12) & 0x0F) | 0xE0; bits= 6; }
4070 { out[i++]=((c >> 18) & 0x07) | 0xF0; bits= 12; }
4072 for ( ; bits >= 0; bits-= 6) {
4073 out[i++]= ((c >> bits) & 0x3F) | 0x80;
4077 htmlCheckParagraph(ctxt);
4078 if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
4079 ctxt->sax->characters(ctxt->userData, out, i);
4086 * @ctxt: an HTML parser context
4088 * Parse a content: comment, sub-element, reference or text.
4089 * Kept for compatibility with old code
4093 htmlParseContent(htmlParserCtxtPtr ctxt) {
4094 xmlChar *currentNode;
4096 const xmlChar *name;
4098 currentNode = xmlStrdup(ctxt->name);
4099 depth = ctxt->nameNr;
4101 long cons = ctxt->nbChars;
4105 if (ctxt->instate == XML_PARSER_EOF)
4109 * Our tag or one of it's parent or children is ending.
4111 if ((CUR == '<') && (NXT(1) == '/')) {
4112 if (htmlParseEndTag(ctxt) &&
4113 ((currentNode != NULL) || (ctxt->nameNr == 0))) {
4114 if (currentNode != NULL)
4115 xmlFree(currentNode);
4118 continue; /* while */
4121 else if ((CUR == '<') &&
4122 ((IS_ASCII_LETTER(NXT(1))) ||
4123 (NXT(1) == '_') || (NXT(1) == ':'))) {
4124 name = htmlParseHTMLName_nonInvasive(ctxt);
4126 htmlParseErr(ctxt, XML_ERR_NAME_REQUIRED,
4127 "htmlParseStartTag: invalid element name\n",
4129 /* Dump the bogus tag like browsers do */
4130 while ((IS_CHAR_CH(CUR)) && (CUR != '>'))
4133 if (currentNode != NULL)
4134 xmlFree(currentNode);
4138 if (ctxt->name != NULL) {
4139 if (htmlCheckAutoClose(name, ctxt->name) == 1) {
4140 htmlAutoClose(ctxt, name);
4147 * Has this node been popped out during parsing of
4150 if ((ctxt->nameNr > 0) && (depth >= ctxt->nameNr) &&
4151 (!xmlStrEqual(currentNode, ctxt->name)))
4153 if (currentNode != NULL) xmlFree(currentNode);
4157 if ((CUR != 0) && ((xmlStrEqual(currentNode, BAD_CAST"script")) ||
4158 (xmlStrEqual(currentNode, BAD_CAST"style")))) {
4160 * Handle SCRIPT/STYLE separately
4162 htmlParseScript(ctxt);
4165 * Sometimes DOCTYPE arrives in the middle of the document
4167 if ((CUR == '<') && (NXT(1) == '!') &&
4168 (UPP(2) == 'D') && (UPP(3) == 'O') &&
4169 (UPP(4) == 'C') && (UPP(5) == 'T') &&
4170 (UPP(6) == 'Y') && (UPP(7) == 'P') &&
4172 htmlParseErr(ctxt, XML_HTML_STRUCURE_ERROR,
4173 "Misplaced DOCTYPE declaration\n",
4174 BAD_CAST "DOCTYPE" , NULL);
4175 htmlParseDocTypeDecl(ctxt);
4179 * First case : a comment
4181 if ((CUR == '<') && (NXT(1) == '!') &&
4182 (NXT(2) == '-') && (NXT(3) == '-')) {
4183 htmlParseComment(ctxt);
4187 * Second case : a Processing Instruction.
4189 else if ((CUR == '<') && (NXT(1) == '?')) {
4194 * Third case : a sub-element.
4196 else if (CUR == '<') {
4197 htmlParseElement(ctxt);
4201 * Fourth case : a reference. If if has not been resolved,
4202 * parsing returns it's Name, create the node
4204 else if (CUR == '&') {
4205 htmlParseReference(ctxt);
4209 * Fifth case : end of the resource
4211 else if (CUR == 0) {
4212 htmlAutoCloseOnEnd(ctxt);
4217 * Last case, text. Note that References are handled directly.
4220 htmlParseCharData(ctxt);
4223 if (cons == ctxt->nbChars) {
4224 if (ctxt->node != NULL) {
4225 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
4226 "detected an error in element content\n",
4234 if (currentNode != NULL) xmlFree(currentNode);
4239 * @ctxt: an HTML parser context
4241 * parse an HTML element, this is highly recursive
4242 * this is kept for compatibility with previous code versions
4244 * [39] element ::= EmptyElemTag | STag content ETag
4246 * [41] Attribute ::= Name Eq AttValue
4250 htmlParseElement(htmlParserCtxtPtr ctxt) {
4251 const xmlChar *name;
4252 xmlChar *currentNode = NULL;
4253 const htmlElemDesc * info;
4254 htmlParserNodeInfo node_info;
4257 const xmlChar *oldptr;
4259 if ((ctxt == NULL) || (ctxt->input == NULL)) {
4260 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
4261 "htmlParseElement: context error\n", NULL, NULL);
4265 if (ctxt->instate == XML_PARSER_EOF)
4268 /* Capture start position */
4269 if (ctxt->record_info) {
4270 node_info.begin_pos = ctxt->input->consumed +
4271 (CUR_PTR - ctxt->input->base);
4272 node_info.begin_line = ctxt->input->line;
4275 failed = htmlParseStartTag(ctxt);
4277 if ((failed == -1) || (name == NULL)) {
4284 * Lookup the info for that element.
4286 info = htmlTagLookup(name);
4288 htmlParseErr(ctxt, XML_HTML_UNKNOWN_TAG,
4289 "Tag %s invalid\n", name, NULL);
4293 * Check for an Empty Element labeled the XML/SGML way
4295 if ((CUR == '/') && (NXT(1) == '>')) {
4297 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
4298 ctxt->sax->endElement(ctxt->userData, name);
4306 htmlParseErr(ctxt, XML_ERR_GT_REQUIRED,
4307 "Couldn't find end of Start Tag %s\n", name, NULL);
4310 * end of parsing of this node.
4312 if (xmlStrEqual(name, ctxt->name)) {
4318 * Capture end position and add node
4320 if (ctxt->record_info) {
4321 node_info.end_pos = ctxt->input->consumed +
4322 (CUR_PTR - ctxt->input->base);
4323 node_info.end_line = ctxt->input->line;
4324 node_info.node = ctxt->node;
4325 xmlParserAddNodeInfo(ctxt, &node_info);
4331 * Check for an Empty Element from DTD definition
4333 if ((info != NULL) && (info->empty)) {
4334 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
4335 ctxt->sax->endElement(ctxt->userData, name);
4341 * Parse the content of the element:
4343 currentNode = xmlStrdup(ctxt->name);
4344 depth = ctxt->nameNr;
4345 while (IS_CHAR_CH(CUR)) {
4346 oldptr = ctxt->input->cur;
4347 htmlParseContent(ctxt);
4348 if (oldptr==ctxt->input->cur) break;
4349 if (ctxt->nameNr < depth) break;
4353 * Capture end position and add node
4355 if ( currentNode != NULL && ctxt->record_info ) {
4356 node_info.end_pos = ctxt->input->consumed +
4357 (CUR_PTR - ctxt->input->base);
4358 node_info.end_line = ctxt->input->line;
4359 node_info.node = ctxt->node;
4360 xmlParserAddNodeInfo(ctxt, &node_info);
4362 if (!IS_CHAR_CH(CUR)) {
4363 htmlAutoCloseOnEnd(ctxt);
4366 if (currentNode != NULL)
4367 xmlFree(currentNode);
4371 htmlParserFinishElementParsing(htmlParserCtxtPtr ctxt) {
4373 * Capture end position and add node
4375 if ( ctxt->node != NULL && ctxt->record_info ) {
4376 ctxt->nodeInfo->end_pos = ctxt->input->consumed +
4377 (CUR_PTR - ctxt->input->base);
4378 ctxt->nodeInfo->end_line = ctxt->input->line;
4379 ctxt->nodeInfo->node = ctxt->node;
4380 xmlParserAddNodeInfo(ctxt, ctxt->nodeInfo);
4381 htmlNodeInfoPop(ctxt);
4383 if (!IS_CHAR_CH(CUR)) {
4384 htmlAutoCloseOnEnd(ctxt);
4389 * htmlParseElementInternal:
4390 * @ctxt: an HTML parser context
4392 * parse an HTML element, new version, non recursive
4394 * [39] element ::= EmptyElemTag | STag content ETag
4396 * [41] Attribute ::= Name Eq AttValue
4400 htmlParseElementInternal(htmlParserCtxtPtr ctxt) {
4401 const xmlChar *name;
4402 const htmlElemDesc * info;
4403 htmlParserNodeInfo node_info = { 0, };
4406 if ((ctxt == NULL) || (ctxt->input == NULL)) {
4407 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
4408 "htmlParseElementInternal: context error\n", NULL, NULL);
4412 if (ctxt->instate == XML_PARSER_EOF)
4415 /* Capture start position */
4416 if (ctxt->record_info) {
4417 node_info.begin_pos = ctxt->input->consumed +
4418 (CUR_PTR - ctxt->input->base);
4419 node_info.begin_line = ctxt->input->line;
4422 failed = htmlParseStartTag(ctxt);
4424 if ((failed == -1) || (name == NULL)) {
4431 * Lookup the info for that element.
4433 info = htmlTagLookup(name);
4435 htmlParseErr(ctxt, XML_HTML_UNKNOWN_TAG,
4436 "Tag %s invalid\n", name, NULL);
4440 * Check for an Empty Element labeled the XML/SGML way
4442 if ((CUR == '/') && (NXT(1) == '>')) {
4444 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
4445 ctxt->sax->endElement(ctxt->userData, name);
4453 htmlParseErr(ctxt, XML_ERR_GT_REQUIRED,
4454 "Couldn't find end of Start Tag %s\n", name, NULL);
4457 * end of parsing of this node.
4459 if (xmlStrEqual(name, ctxt->name)) {
4464 if (ctxt->record_info)
4465 htmlNodeInfoPush(ctxt, &node_info);
4466 htmlParserFinishElementParsing(ctxt);
4471 * Check for an Empty Element from DTD definition
4473 if ((info != NULL) && (info->empty)) {
4474 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
4475 ctxt->sax->endElement(ctxt->userData, name);
4480 if (ctxt->record_info)
4481 htmlNodeInfoPush(ctxt, &node_info);
4485 * htmlParseContentInternal:
4486 * @ctxt: an HTML parser context
4488 * Parse a content: comment, sub-element, reference or text.
4489 * New version for non recursive htmlParseElementInternal
4493 htmlParseContentInternal(htmlParserCtxtPtr ctxt) {
4494 xmlChar *currentNode;
4496 const xmlChar *name;
4498 currentNode = xmlStrdup(ctxt->name);
4499 depth = ctxt->nameNr;
4501 long cons = ctxt->nbChars;
4505 if (ctxt->instate == XML_PARSER_EOF)
4509 * Our tag or one of it's parent or children is ending.
4511 if ((CUR == '<') && (NXT(1) == '/')) {
4512 if (htmlParseEndTag(ctxt) &&
4513 ((currentNode != NULL) || (ctxt->nameNr == 0))) {
4514 if (currentNode != NULL)
4515 xmlFree(currentNode);
4517 currentNode = xmlStrdup(ctxt->name);
4518 depth = ctxt->nameNr;
4520 continue; /* while */
4523 else if ((CUR == '<') &&
4524 ((IS_ASCII_LETTER(NXT(1))) ||
4525 (NXT(1) == '_') || (NXT(1) == ':'))) {
4526 name = htmlParseHTMLName_nonInvasive(ctxt);
4528 htmlParseErr(ctxt, XML_ERR_NAME_REQUIRED,
4529 "htmlParseStartTag: invalid element name\n",
4531 /* Dump the bogus tag like browsers do */
4532 while ((IS_CHAR_CH(CUR)) && (CUR != '>'))
4535 htmlParserFinishElementParsing(ctxt);
4536 if (currentNode != NULL)
4537 xmlFree(currentNode);
4539 currentNode = xmlStrdup(ctxt->name);
4540 depth = ctxt->nameNr;
4544 if (ctxt->name != NULL) {
4545 if (htmlCheckAutoClose(name, ctxt->name) == 1) {
4546 htmlAutoClose(ctxt, name);
4553 * Has this node been popped out during parsing of
4556 if ((ctxt->nameNr > 0) && (depth >= ctxt->nameNr) &&
4557 (!xmlStrEqual(currentNode, ctxt->name)))
4559 htmlParserFinishElementParsing(ctxt);
4560 if (currentNode != NULL) xmlFree(currentNode);
4562 currentNode = xmlStrdup(ctxt->name);
4563 depth = ctxt->nameNr;
4567 if ((CUR != 0) && ((xmlStrEqual(currentNode, BAD_CAST"script")) ||
4568 (xmlStrEqual(currentNode, BAD_CAST"style")))) {
4570 * Handle SCRIPT/STYLE separately
4572 htmlParseScript(ctxt);
4575 * Sometimes DOCTYPE arrives in the middle of the document
4577 if ((CUR == '<') && (NXT(1) == '!') &&
4578 (UPP(2) == 'D') && (UPP(3) == 'O') &&
4579 (UPP(4) == 'C') && (UPP(5) == 'T') &&
4580 (UPP(6) == 'Y') && (UPP(7) == 'P') &&
4582 htmlParseErr(ctxt, XML_HTML_STRUCURE_ERROR,
4583 "Misplaced DOCTYPE declaration\n",
4584 BAD_CAST "DOCTYPE" , NULL);
4585 htmlParseDocTypeDecl(ctxt);
4589 * First case : a comment
4591 if ((CUR == '<') && (NXT(1) == '!') &&
4592 (NXT(2) == '-') && (NXT(3) == '-')) {
4593 htmlParseComment(ctxt);
4597 * Second case : a Processing Instruction.
4599 else if ((CUR == '<') && (NXT(1) == '?')) {
4604 * Third case : a sub-element.
4606 else if (CUR == '<') {
4607 htmlParseElementInternal(ctxt);
4608 if (currentNode != NULL) xmlFree(currentNode);
4610 currentNode = xmlStrdup(ctxt->name);
4611 depth = ctxt->nameNr;
4615 * Fourth case : a reference. If if has not been resolved,
4616 * parsing returns it's Name, create the node
4618 else if (CUR == '&') {
4619 htmlParseReference(ctxt);
4623 * Fifth case : end of the resource
4625 else if (CUR == 0) {
4626 htmlAutoCloseOnEnd(ctxt);
4631 * Last case, text. Note that References are handled directly.
4634 htmlParseCharData(ctxt);
4637 if (cons == ctxt->nbChars) {
4638 if (ctxt->node != NULL) {
4639 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
4640 "detected an error in element content\n",
4648 if (currentNode != NULL) xmlFree(currentNode);
4653 * @ctxt: an HTML parser context
4655 * Parse a content: comment, sub-element, reference or text.
4656 * This is the entry point when called from parser.c
4660 __htmlParseContent(void *ctxt) {
4662 htmlParseContentInternal((htmlParserCtxtPtr) ctxt);
4666 * htmlParseDocument:
4667 * @ctxt: an HTML parser context
4669 * parse an HTML document (and build a tree if using the standard SAX
4672 * Returns 0, -1 in case of error. the parser context is augmented
4673 * as a result of the parsing.
4677 htmlParseDocument(htmlParserCtxtPtr ctxt) {
4679 xmlCharEncoding enc;
4684 htmlDefaultSAXHandlerInit();
4686 if ((ctxt == NULL) || (ctxt->input == NULL)) {
4687 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
4688 "htmlParseDocument: context error\n", NULL, NULL);
4689 return(XML_ERR_INTERNAL_ERROR);
4692 ctxt->linenumbers = 1;
4695 * SAX: beginning of the document processing.
4697 if ((ctxt->sax) && (ctxt->sax->setDocumentLocator))
4698 ctxt->sax->setDocumentLocator(ctxt->userData, &xmlDefaultSAXLocator);
4700 if ((ctxt->encoding == (const xmlChar *)XML_CHAR_ENCODING_NONE) &&
4701 ((ctxt->input->end - ctxt->input->cur) >= 4)) {
4703 * Get the 4 first bytes and decode the charset
4704 * if enc != XML_CHAR_ENCODING_NONE
4705 * plug some encoding conversion routines.
4711 enc = xmlDetectCharEncoding(&start[0], 4);
4712 if (enc != XML_CHAR_ENCODING_NONE) {
4713 xmlSwitchEncoding(ctxt, enc);
4718 * Wipe out everything which is before the first '<'
4722 htmlParseErr(ctxt, XML_ERR_DOCUMENT_EMPTY,
4723 "Document is empty\n", NULL, NULL);
4726 if ((ctxt->sax) && (ctxt->sax->startDocument) && (!ctxt->disableSAX))
4727 ctxt->sax->startDocument(ctxt->userData);
4731 * Parse possible comments and PIs before any content
4733 while (((CUR == '<') && (NXT(1) == '!') &&
4734 (NXT(2) == '-') && (NXT(3) == '-')) ||
4735 ((CUR == '<') && (NXT(1) == '?'))) {
4736 htmlParseComment(ctxt);
4743 * Then possibly doc type declaration(s) and more Misc
4744 * (doctypedecl Misc*)?
4746 if ((CUR == '<') && (NXT(1) == '!') &&
4747 (UPP(2) == 'D') && (UPP(3) == 'O') &&
4748 (UPP(4) == 'C') && (UPP(5) == 'T') &&
4749 (UPP(6) == 'Y') && (UPP(7) == 'P') &&
4751 htmlParseDocTypeDecl(ctxt);
4756 * Parse possible comments and PIs before any content
4758 while (((CUR == '<') && (NXT(1) == '!') &&
4759 (NXT(2) == '-') && (NXT(3) == '-')) ||
4760 ((CUR == '<') && (NXT(1) == '?'))) {
4761 htmlParseComment(ctxt);
4767 * Time to start parsing the tree itself
4769 htmlParseContentInternal(ctxt);
4775 htmlAutoCloseOnEnd(ctxt);
4779 * SAX: end of the document processing.
4781 if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
4782 ctxt->sax->endDocument(ctxt->userData);
4784 if ((!(ctxt->options & HTML_PARSE_NODEFDTD)) && (ctxt->myDoc != NULL)) {
4785 dtd = xmlGetIntSubset(ctxt->myDoc);
4787 ctxt->myDoc->intSubset =
4788 xmlCreateIntSubset(ctxt->myDoc, BAD_CAST "html",
4789 BAD_CAST "-//W3C//DTD HTML 4.0 Transitional//EN",
4790 BAD_CAST "http://www.w3.org/TR/REC-html40/loose.dtd");
4792 if (! ctxt->wellFormed) return(-1);
4797 /************************************************************************
4799 * Parser contexts handling *
4801 ************************************************************************/
4804 * htmlInitParserCtxt:
4805 * @ctxt: an HTML parser context
4807 * Initialize a parser context
4809 * Returns 0 in case of success and -1 in case of error
4813 htmlInitParserCtxt(htmlParserCtxtPtr ctxt)
4815 htmlSAXHandler *sax;
4817 if (ctxt == NULL) return(-1);
4818 memset(ctxt, 0, sizeof(htmlParserCtxt));
4820 ctxt->dict = xmlDictCreate();
4821 if (ctxt->dict == NULL) {
4822 htmlErrMemory(NULL, "htmlInitParserCtxt: out of memory\n");
4825 sax = (htmlSAXHandler *) xmlMalloc(sizeof(htmlSAXHandler));
4827 htmlErrMemory(NULL, "htmlInitParserCtxt: out of memory\n");
4831 memset(sax, 0, sizeof(htmlSAXHandler));
4833 /* Allocate the Input stack */
4834 ctxt->inputTab = (htmlParserInputPtr *)
4835 xmlMalloc(5 * sizeof(htmlParserInputPtr));
4836 if (ctxt->inputTab == NULL) {
4837 htmlErrMemory(NULL, "htmlInitParserCtxt: out of memory\n");
4846 ctxt->version = NULL;
4847 ctxt->encoding = NULL;
4848 ctxt->standalone = -1;
4849 ctxt->instate = XML_PARSER_START;
4851 /* Allocate the Node stack */
4852 ctxt->nodeTab = (htmlNodePtr *) xmlMalloc(10 * sizeof(htmlNodePtr));
4853 if (ctxt->nodeTab == NULL) {
4854 htmlErrMemory(NULL, "htmlInitParserCtxt: out of memory\n");
4867 /* Allocate the Name stack */
4868 ctxt->nameTab = (const xmlChar **) xmlMalloc(10 * sizeof(xmlChar *));
4869 if (ctxt->nameTab == NULL) {
4870 htmlErrMemory(NULL, "htmlInitParserCtxt: out of memory\n");
4886 ctxt->nodeInfoTab = NULL;
4887 ctxt->nodeInfoNr = 0;
4888 ctxt->nodeInfoMax = 0;
4890 if (sax == NULL) ctxt->sax = (xmlSAXHandlerPtr) &htmlDefaultSAXHandler;
4893 memcpy(sax, &htmlDefaultSAXHandler, sizeof(xmlSAXHandlerV1));
4895 ctxt->userData = ctxt;
4897 ctxt->wellFormed = 1;
4898 ctxt->replaceEntities = 0;
4899 ctxt->linenumbers = xmlLineNumbersDefaultValue;
4901 ctxt->vctxt.finishDtd = XML_CTXT_FINISH_DTD_0;
4902 ctxt->vctxt.userData = ctxt;
4903 ctxt->vctxt.error = xmlParserValidityError;
4904 ctxt->vctxt.warning = xmlParserValidityWarning;
4905 ctxt->record_info = 0;
4908 ctxt->checkIndex = 0;
4909 ctxt->catalogs = NULL;
4910 xmlInitNodeInfoSeq(&ctxt->node_seq);
4915 * htmlFreeParserCtxt:
4916 * @ctxt: an HTML parser context
4918 * Free all the memory used by a parser context. However the parsed
4919 * document in ctxt->myDoc is not freed.
4923 htmlFreeParserCtxt(htmlParserCtxtPtr ctxt)
4925 xmlFreeParserCtxt(ctxt);
4929 * htmlNewParserCtxt:
4931 * Allocate and initialize a new parser context.
4933 * Returns the htmlParserCtxtPtr or NULL in case of allocation error
4937 htmlNewParserCtxt(void)
4939 xmlParserCtxtPtr ctxt;
4941 ctxt = (xmlParserCtxtPtr) xmlMalloc(sizeof(xmlParserCtxt));
4943 htmlErrMemory(NULL, "NewParserCtxt: out of memory\n");
4946 memset(ctxt, 0, sizeof(xmlParserCtxt));
4947 if (htmlInitParserCtxt(ctxt) < 0) {
4948 htmlFreeParserCtxt(ctxt);
4955 * htmlCreateMemoryParserCtxt:
4956 * @buffer: a pointer to a char array
4957 * @size: the size of the array
4959 * Create a parser context for an HTML in-memory document.
4961 * Returns the new parser context or NULL
4964 htmlCreateMemoryParserCtxt(const char *buffer, int size) {
4965 xmlParserCtxtPtr ctxt;
4966 xmlParserInputPtr input;
4967 xmlParserInputBufferPtr buf;
4974 ctxt = htmlNewParserCtxt();
4978 buf = xmlParserInputBufferCreateMem(buffer, size, XML_CHAR_ENCODING_NONE);
4979 if (buf == NULL) return(NULL);
4981 input = xmlNewInputStream(ctxt);
4982 if (input == NULL) {
4983 xmlFreeParserCtxt(ctxt);
4987 input->filename = NULL;
4989 xmlBufResetInput(buf->buffer, input);
4991 inputPush(ctxt, input);
4996 * htmlCreateDocParserCtxt:
4997 * @cur: a pointer to an array of xmlChar
4998 * @encoding: a free form C string describing the HTML document encoding, or NULL
5000 * Create a parser context for an HTML document.
5002 * TODO: check the need to add encoding handling there
5004 * Returns the new parser context or NULL
5006 static htmlParserCtxtPtr
5007 htmlCreateDocParserCtxt(const xmlChar *cur, const char *encoding) {
5009 htmlParserCtxtPtr ctxt;
5013 len = xmlStrlen(cur);
5014 ctxt = htmlCreateMemoryParserCtxt((char *)cur, len);
5018 if (encoding != NULL) {
5019 xmlCharEncoding enc;
5020 xmlCharEncodingHandlerPtr handler;
5022 if (ctxt->input->encoding != NULL)
5023 xmlFree((xmlChar *) ctxt->input->encoding);
5024 ctxt->input->encoding = xmlStrdup((const xmlChar *) encoding);
5026 enc = xmlParseCharEncoding(encoding);
5028 * registered set of known encodings
5030 if (enc != XML_CHAR_ENCODING_ERROR) {
5031 xmlSwitchEncoding(ctxt, enc);
5032 if (ctxt->errNo == XML_ERR_UNSUPPORTED_ENCODING) {
5033 htmlParseErr(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
5034 "Unsupported encoding %s\n",
5035 (const xmlChar *) encoding, NULL);
5039 * fallback for unknown encodings
5041 handler = xmlFindCharEncodingHandler((const char *) encoding);
5042 if (handler != NULL) {
5043 xmlSwitchToEncoding(ctxt, handler);
5045 htmlParseErr(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
5046 "Unsupported encoding %s\n",
5047 (const xmlChar *) encoding, NULL);
5054 #ifdef LIBXML_PUSH_ENABLED
5055 /************************************************************************
5057 * Progressive parsing interfaces *
5059 ************************************************************************/
5062 * htmlParseLookupSequence:
5063 * @ctxt: an HTML parser context
5064 * @first: the first char to lookup
5065 * @next: the next char to lookup or zero
5066 * @third: the next char to lookup or zero
5067 * @comment: flag to force checking inside comments
5069 * Try to find if a sequence (first, next, third) or just (first next) or
5070 * (first) is available in the input stream.
5071 * This function has a side effect of (possibly) incrementing ctxt->checkIndex
5072 * to avoid rescanning sequences of bytes, it DOES change the state of the
5073 * parser, do not use liberally.
5074 * This is basically similar to xmlParseLookupSequence()
5076 * Returns the index to the current parsing point if the full sequence
5077 * is available, -1 otherwise.
5080 htmlParseLookupSequence(htmlParserCtxtPtr ctxt, xmlChar first,
5081 xmlChar next, xmlChar third, int iscomment,
5085 htmlParserInputPtr in;
5089 char valdellim = 0x0;
5095 base = in->cur - in->base;
5099 if (ctxt->checkIndex > base)
5100 base = ctxt->checkIndex;
5102 if (in->buf == NULL) {
5106 buf = xmlBufContent(in->buf->buffer);
5107 len = xmlBufUse(in->buf->buffer);
5110 /* take into account the sequence length */
5115 for (; base < len; base++) {
5116 if ((!incomment) && (base + 4 < len) && (!iscomment)) {
5117 if ((buf[base] == '<') && (buf[base + 1] == '!') &&
5118 (buf[base + 2] == '-') && (buf[base + 3] == '-')) {
5120 /* do not increment past <! - some people use <!--> */
5124 if (ignoreattrval) {
5125 if (buf[base] == '"' || buf[base] == '\'') {
5127 if (buf[base] == valdellim) {
5132 valdellim = buf[base];
5136 } else if (invalue) {
5143 if ((buf[base] == '-') && (buf[base + 1] == '-') &&
5144 (buf[base + 2] == '>')) {
5150 if (buf[base] == first) {
5152 if ((buf[base + 1] != next) || (buf[base + 2] != third))
5154 } else if (next != 0) {
5155 if (buf[base + 1] != next)
5158 ctxt->checkIndex = 0;
5161 xmlGenericError(xmlGenericErrorContext,
5162 "HPP: lookup '%c' found at %d\n",
5164 else if (third == 0)
5165 xmlGenericError(xmlGenericErrorContext,
5166 "HPP: lookup '%c%c' found at %d\n",
5169 xmlGenericError(xmlGenericErrorContext,
5170 "HPP: lookup '%c%c%c' found at %d\n",
5171 first, next, third, base);
5173 return (base - (in->cur - in->base));
5176 if ((!incomment) && (!invalue))
5177 ctxt->checkIndex = base;
5180 xmlGenericError(xmlGenericErrorContext,
5181 "HPP: lookup '%c' failed\n", first);
5182 else if (third == 0)
5183 xmlGenericError(xmlGenericErrorContext,
5184 "HPP: lookup '%c%c' failed\n", first, next);
5186 xmlGenericError(xmlGenericErrorContext,
5187 "HPP: lookup '%c%c%c' failed\n", first, next,
5194 * htmlParseLookupChars:
5195 * @ctxt: an HTML parser context
5196 * @stop: Array of chars, which stop the lookup.
5197 * @stopLen: Length of stop-Array
5199 * Try to find if any char of the stop-Array is available in the input
5201 * This function has a side effect of (possibly) incrementing ctxt->checkIndex
5202 * to avoid rescanning sequences of bytes, it DOES change the state of the
5203 * parser, do not use liberally.
5205 * Returns the index to the current parsing point if a stopChar
5206 * is available, -1 otherwise.
5209 htmlParseLookupChars(htmlParserCtxtPtr ctxt, const xmlChar * stop,
5213 htmlParserInputPtr in;
5222 base = in->cur - in->base;
5226 if (ctxt->checkIndex > base)
5227 base = ctxt->checkIndex;
5229 if (in->buf == NULL) {
5233 buf = xmlBufContent(in->buf->buffer);
5234 len = xmlBufUse(in->buf->buffer);
5237 for (; base < len; base++) {
5238 if (!incomment && (base + 4 < len)) {
5239 if ((buf[base] == '<') && (buf[base + 1] == '!') &&
5240 (buf[base + 2] == '-') && (buf[base + 3] == '-')) {
5242 /* do not increment past <! - some people use <!--> */
5249 if ((buf[base] == '-') && (buf[base + 1] == '-') &&
5250 (buf[base + 2] == '>')) {
5256 for (i = 0; i < stopLen; ++i) {
5257 if (buf[base] == stop[i]) {
5258 ctxt->checkIndex = 0;
5259 return (base - (in->cur - in->base));
5263 ctxt->checkIndex = base;
5268 * htmlParseTryOrFinish:
5269 * @ctxt: an HTML parser context
5270 * @terminate: last chunk indicator
5272 * Try to progress on parsing
5274 * Returns zero if no parsing was possible
5277 htmlParseTryOrFinish(htmlParserCtxtPtr ctxt, int terminate) {
5279 htmlParserInputPtr in;
5283 htmlParserNodeInfo node_info;
5286 switch (ctxt->instate) {
5287 case XML_PARSER_EOF:
5288 xmlGenericError(xmlGenericErrorContext,
5289 "HPP: try EOF\n"); break;
5290 case XML_PARSER_START:
5291 xmlGenericError(xmlGenericErrorContext,
5292 "HPP: try START\n"); break;
5293 case XML_PARSER_MISC:
5294 xmlGenericError(xmlGenericErrorContext,
5295 "HPP: try MISC\n");break;
5296 case XML_PARSER_COMMENT:
5297 xmlGenericError(xmlGenericErrorContext,
5298 "HPP: try COMMENT\n");break;
5299 case XML_PARSER_PROLOG:
5300 xmlGenericError(xmlGenericErrorContext,
5301 "HPP: try PROLOG\n");break;
5302 case XML_PARSER_START_TAG:
5303 xmlGenericError(xmlGenericErrorContext,
5304 "HPP: try START_TAG\n");break;
5305 case XML_PARSER_CONTENT:
5306 xmlGenericError(xmlGenericErrorContext,
5307 "HPP: try CONTENT\n");break;
5308 case XML_PARSER_CDATA_SECTION:
5309 xmlGenericError(xmlGenericErrorContext,
5310 "HPP: try CDATA_SECTION\n");break;
5311 case XML_PARSER_END_TAG:
5312 xmlGenericError(xmlGenericErrorContext,
5313 "HPP: try END_TAG\n");break;
5314 case XML_PARSER_ENTITY_DECL:
5315 xmlGenericError(xmlGenericErrorContext,
5316 "HPP: try ENTITY_DECL\n");break;
5317 case XML_PARSER_ENTITY_VALUE:
5318 xmlGenericError(xmlGenericErrorContext,
5319 "HPP: try ENTITY_VALUE\n");break;
5320 case XML_PARSER_ATTRIBUTE_VALUE:
5321 xmlGenericError(xmlGenericErrorContext,
5322 "HPP: try ATTRIBUTE_VALUE\n");break;
5323 case XML_PARSER_DTD:
5324 xmlGenericError(xmlGenericErrorContext,
5325 "HPP: try DTD\n");break;
5326 case XML_PARSER_EPILOG:
5327 xmlGenericError(xmlGenericErrorContext,
5328 "HPP: try EPILOG\n");break;
5330 xmlGenericError(xmlGenericErrorContext,
5331 "HPP: try PI\n");break;
5332 case XML_PARSER_SYSTEM_LITERAL:
5333 xmlGenericError(xmlGenericErrorContext,
5334 "HPP: try SYSTEM_LITERAL\n");break;
5341 if (in == NULL) break;
5342 if (in->buf == NULL)
5343 avail = in->length - (in->cur - in->base);
5345 avail = xmlBufUse(in->buf->buffer) - (in->cur - in->base);
5346 if ((avail == 0) && (terminate)) {
5347 htmlAutoCloseOnEnd(ctxt);
5348 if ((ctxt->nameNr == 0) && (ctxt->instate != XML_PARSER_EOF)) {
5350 * SAX: end of the document processing.
5352 ctxt->instate = XML_PARSER_EOF;
5353 if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
5354 ctxt->sax->endDocument(ctxt->userData);
5365 switch (ctxt->instate) {
5366 case XML_PARSER_EOF:
5368 * Document parsing is done !
5371 case XML_PARSER_START:
5373 * Very first chars read from the document flow.
5376 if (IS_BLANK_CH(cur)) {
5378 if (in->buf == NULL)
5379 avail = in->length - (in->cur - in->base);
5381 avail = xmlBufUse(in->buf->buffer) - (in->cur - in->base);
5383 if ((ctxt->sax) && (ctxt->sax->setDocumentLocator))
5384 ctxt->sax->setDocumentLocator(ctxt->userData,
5385 &xmlDefaultSAXLocator);
5386 if ((ctxt->sax) && (ctxt->sax->startDocument) &&
5387 (!ctxt->disableSAX))
5388 ctxt->sax->startDocument(ctxt->userData);
5392 if ((cur == '<') && (next == '!') &&
5393 (UPP(2) == 'D') && (UPP(3) == 'O') &&
5394 (UPP(4) == 'C') && (UPP(5) == 'T') &&
5395 (UPP(6) == 'Y') && (UPP(7) == 'P') &&
5398 (htmlParseLookupSequence(ctxt, '>', 0, 0, 0, 1) < 0))
5401 xmlGenericError(xmlGenericErrorContext,
5402 "HPP: Parsing internal subset\n");
5404 htmlParseDocTypeDecl(ctxt);
5405 ctxt->instate = XML_PARSER_PROLOG;
5407 xmlGenericError(xmlGenericErrorContext,
5408 "HPP: entering PROLOG\n");
5411 ctxt->instate = XML_PARSER_MISC;
5413 xmlGenericError(xmlGenericErrorContext,
5414 "HPP: entering MISC\n");
5418 case XML_PARSER_MISC:
5420 if (in->buf == NULL)
5421 avail = in->length - (in->cur - in->base);
5423 avail = xmlBufUse(in->buf->buffer) - (in->cur - in->base);
5425 * no chars in buffer
5430 * not enouth chars in buffer
5441 if ((cur == '<') && (next == '!') &&
5442 (in->cur[2] == '-') && (in->cur[3] == '-')) {
5444 (htmlParseLookupSequence(ctxt, '-', '-', '>', 1, 1) < 0))
5447 xmlGenericError(xmlGenericErrorContext,
5448 "HPP: Parsing Comment\n");
5450 htmlParseComment(ctxt);
5451 ctxt->instate = XML_PARSER_MISC;
5452 } else if ((cur == '<') && (next == '?')) {
5454 (htmlParseLookupSequence(ctxt, '>', 0, 0, 0, 1) < 0))
5457 xmlGenericError(xmlGenericErrorContext,
5458 "HPP: Parsing PI\n");
5461 ctxt->instate = XML_PARSER_MISC;
5462 } else if ((cur == '<') && (next == '!') &&
5463 (UPP(2) == 'D') && (UPP(3) == 'O') &&
5464 (UPP(4) == 'C') && (UPP(5) == 'T') &&
5465 (UPP(6) == 'Y') && (UPP(7) == 'P') &&
5468 (htmlParseLookupSequence(ctxt, '>', 0, 0, 0, 1) < 0))
5471 xmlGenericError(xmlGenericErrorContext,
5472 "HPP: Parsing internal subset\n");
5474 htmlParseDocTypeDecl(ctxt);
5475 ctxt->instate = XML_PARSER_PROLOG;
5477 xmlGenericError(xmlGenericErrorContext,
5478 "HPP: entering PROLOG\n");
5480 } else if ((cur == '<') && (next == '!') &&
5484 ctxt->instate = XML_PARSER_START_TAG;
5486 xmlGenericError(xmlGenericErrorContext,
5487 "HPP: entering START_TAG\n");
5491 case XML_PARSER_PROLOG:
5493 if (in->buf == NULL)
5494 avail = in->length - (in->cur - in->base);
5496 avail = xmlBufUse(in->buf->buffer) - (in->cur - in->base);
5501 if ((cur == '<') && (next == '!') &&
5502 (in->cur[2] == '-') && (in->cur[3] == '-')) {
5504 (htmlParseLookupSequence(ctxt, '-', '-', '>', 1, 1) < 0))
5507 xmlGenericError(xmlGenericErrorContext,
5508 "HPP: Parsing Comment\n");
5510 htmlParseComment(ctxt);
5511 ctxt->instate = XML_PARSER_PROLOG;
5512 } else if ((cur == '<') && (next == '?')) {
5514 (htmlParseLookupSequence(ctxt, '>', 0, 0, 0, 1) < 0))
5517 xmlGenericError(xmlGenericErrorContext,
5518 "HPP: Parsing PI\n");
5521 ctxt->instate = XML_PARSER_PROLOG;
5522 } else if ((cur == '<') && (next == '!') &&
5526 ctxt->instate = XML_PARSER_START_TAG;
5528 xmlGenericError(xmlGenericErrorContext,
5529 "HPP: entering START_TAG\n");
5533 case XML_PARSER_EPILOG:
5534 if (in->buf == NULL)
5535 avail = in->length - (in->cur - in->base);
5537 avail = xmlBufUse(in->buf->buffer) - (in->cur - in->base);
5541 if (IS_BLANK_CH(cur)) {
5542 htmlParseCharData(ctxt);
5548 if ((cur == '<') && (next == '!') &&
5549 (in->cur[2] == '-') && (in->cur[3] == '-')) {
5551 (htmlParseLookupSequence(ctxt, '-', '-', '>', 1, 1) < 0))
5554 xmlGenericError(xmlGenericErrorContext,
5555 "HPP: Parsing Comment\n");
5557 htmlParseComment(ctxt);
5558 ctxt->instate = XML_PARSER_EPILOG;
5559 } else if ((cur == '<') && (next == '?')) {
5561 (htmlParseLookupSequence(ctxt, '>', 0, 0, 0, 1) < 0))
5564 xmlGenericError(xmlGenericErrorContext,
5565 "HPP: Parsing PI\n");
5568 ctxt->instate = XML_PARSER_EPILOG;
5569 } else if ((cur == '<') && (next == '!') &&
5573 ctxt->errNo = XML_ERR_DOCUMENT_END;
5574 ctxt->wellFormed = 0;
5575 ctxt->instate = XML_PARSER_EOF;
5577 xmlGenericError(xmlGenericErrorContext,
5578 "HPP: entering EOF\n");
5580 if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
5581 ctxt->sax->endDocument(ctxt->userData);
5585 case XML_PARSER_START_TAG: {
5586 const xmlChar *name;
5588 const htmlElemDesc * info;
5591 * no chars in buffer
5596 * not enouth chars in buffer
5608 ctxt->instate = XML_PARSER_CONTENT;
5610 xmlGenericError(xmlGenericErrorContext,
5611 "HPP: entering CONTENT\n");
5616 ctxt->instate = XML_PARSER_END_TAG;
5617 ctxt->checkIndex = 0;
5619 xmlGenericError(xmlGenericErrorContext,
5620 "HPP: entering END_TAG\n");
5625 (htmlParseLookupSequence(ctxt, '>', 0, 0, 0, 1) < 0))
5628 /* Capture start position */
5629 if (ctxt->record_info) {
5630 node_info.begin_pos = ctxt->input->consumed +
5631 (CUR_PTR - ctxt->input->base);
5632 node_info.begin_line = ctxt->input->line;
5636 failed = htmlParseStartTag(ctxt);
5638 if ((failed == -1) ||
5646 * Lookup the info for that element.
5648 info = htmlTagLookup(name);
5650 htmlParseErr(ctxt, XML_HTML_UNKNOWN_TAG,
5651 "Tag %s invalid\n", name, NULL);
5655 * Check for an Empty Element labeled the XML/SGML way
5657 if ((CUR == '/') && (NXT(1) == '>')) {
5659 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
5660 ctxt->sax->endElement(ctxt->userData, name);
5662 ctxt->instate = XML_PARSER_CONTENT;
5664 xmlGenericError(xmlGenericErrorContext,
5665 "HPP: entering CONTENT\n");
5673 htmlParseErr(ctxt, XML_ERR_GT_REQUIRED,
5674 "Couldn't find end of Start Tag %s\n",
5678 * end of parsing of this node.
5680 if (xmlStrEqual(name, ctxt->name)) {
5685 if (ctxt->record_info)
5686 htmlNodeInfoPush(ctxt, &node_info);
5688 ctxt->instate = XML_PARSER_CONTENT;
5690 xmlGenericError(xmlGenericErrorContext,
5691 "HPP: entering CONTENT\n");
5697 * Check for an Empty Element from DTD definition
5699 if ((info != NULL) && (info->empty)) {
5700 if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
5701 ctxt->sax->endElement(ctxt->userData, name);
5705 if (ctxt->record_info)
5706 htmlNodeInfoPush(ctxt, &node_info);
5708 ctxt->instate = XML_PARSER_CONTENT;
5710 xmlGenericError(xmlGenericErrorContext,
5711 "HPP: entering CONTENT\n");
5715 case XML_PARSER_CONTENT: {
5718 * Handle preparsed entities and charRef
5720 if (ctxt->token != 0) {
5721 xmlChar chr[2] = { 0 , 0 } ;
5723 chr[0] = (xmlChar) ctxt->token;
5724 htmlCheckParagraph(ctxt);
5725 if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
5726 ctxt->sax->characters(ctxt->userData, chr, 1);
5728 ctxt->checkIndex = 0;
5730 if ((avail == 1) && (terminate)) {
5732 if ((cur != '<') && (cur != '&')) {
5733 if (ctxt->sax != NULL) {
5734 if (IS_BLANK_CH(cur)) {
5735 if (ctxt->keepBlanks) {
5736 if (ctxt->sax->characters != NULL)
5737 ctxt->sax->characters(
5738 ctxt->userData, &in->cur[0], 1);
5740 if (ctxt->sax->ignorableWhitespace != NULL)
5741 ctxt->sax->ignorableWhitespace(
5742 ctxt->userData, &in->cur[0], 1);
5745 htmlCheckParagraph(ctxt);
5746 if (ctxt->sax->characters != NULL)
5747 ctxt->sax->characters(
5748 ctxt->userData, &in->cur[0], 1);
5752 ctxt->checkIndex = 0;
5761 cons = ctxt->nbChars;
5762 if ((xmlStrEqual(ctxt->name, BAD_CAST"script")) ||
5763 (xmlStrEqual(ctxt->name, BAD_CAST"style"))) {
5765 * Handle SCRIPT/STYLE separately
5771 idx = htmlParseLookupSequence(ctxt, '<', '/', 0, 0, 0);
5774 val = in->cur[idx + 2];
5775 if (val == 0) /* bad cut of input */
5778 htmlParseScript(ctxt);
5779 if ((cur == '<') && (next == '/')) {
5780 ctxt->instate = XML_PARSER_END_TAG;
5781 ctxt->checkIndex = 0;
5783 xmlGenericError(xmlGenericErrorContext,
5784 "HPP: entering END_TAG\n");
5790 * Sometimes DOCTYPE arrives in the middle of the document
5792 if ((cur == '<') && (next == '!') &&
5793 (UPP(2) == 'D') && (UPP(3) == 'O') &&
5794 (UPP(4) == 'C') && (UPP(5) == 'T') &&
5795 (UPP(6) == 'Y') && (UPP(7) == 'P') &&
5798 (htmlParseLookupSequence(ctxt, '>', 0, 0, 0, 1) < 0))
5800 htmlParseErr(ctxt, XML_HTML_STRUCURE_ERROR,
5801 "Misplaced DOCTYPE declaration\n",
5802 BAD_CAST "DOCTYPE" , NULL);
5803 htmlParseDocTypeDecl(ctxt);
5804 } else if ((cur == '<') && (next == '!') &&
5805 (in->cur[2] == '-') && (in->cur[3] == '-')) {
5807 (htmlParseLookupSequence(
5808 ctxt, '-', '-', '>', 1, 1) < 0))
5811 xmlGenericError(xmlGenericErrorContext,
5812 "HPP: Parsing Comment\n");
5814 htmlParseComment(ctxt);
5815 ctxt->instate = XML_PARSER_CONTENT;
5816 } else if ((cur == '<') && (next == '?')) {
5818 (htmlParseLookupSequence(ctxt, '>', 0, 0, 0, 1) < 0))
5821 xmlGenericError(xmlGenericErrorContext,
5822 "HPP: Parsing PI\n");
5825 ctxt->instate = XML_PARSER_CONTENT;
5826 } else if ((cur == '<') && (next == '!') && (avail < 4)) {
5828 } else if ((cur == '<') && (next == '/')) {
5829 ctxt->instate = XML_PARSER_END_TAG;
5830 ctxt->checkIndex = 0;
5832 xmlGenericError(xmlGenericErrorContext,
5833 "HPP: entering END_TAG\n");
5836 } else if (cur == '<') {
5837 ctxt->instate = XML_PARSER_START_TAG;
5838 ctxt->checkIndex = 0;
5840 xmlGenericError(xmlGenericErrorContext,
5841 "HPP: entering START_TAG\n");
5844 } else if (cur == '&') {
5846 (htmlParseLookupChars(ctxt,
5847 BAD_CAST "; >/", 4) < 0))
5850 xmlGenericError(xmlGenericErrorContext,
5851 "HPP: Parsing Reference\n");
5853 /* TODO: check generation of subtrees if noent !!! */
5854 htmlParseReference(ctxt);
5857 * check that the text sequence is complete
5858 * before handing out the data to the parser
5859 * to avoid problems with erroneous end of
5863 (htmlParseLookupChars(ctxt, BAD_CAST "<&", 2) < 0))
5865 ctxt->checkIndex = 0;
5867 xmlGenericError(xmlGenericErrorContext,
5868 "HPP: Parsing char data\n");
5870 htmlParseCharData(ctxt);
5873 if (cons == ctxt->nbChars) {
5874 if (ctxt->node != NULL) {
5875 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
5876 "detected an error in element content\n",
5885 case XML_PARSER_END_TAG:
5889 (htmlParseLookupSequence(ctxt, '>', 0, 0, 0, 1) < 0))
5891 htmlParseEndTag(ctxt);
5892 if (ctxt->nameNr == 0) {
5893 ctxt->instate = XML_PARSER_EPILOG;
5895 ctxt->instate = XML_PARSER_CONTENT;
5897 ctxt->checkIndex = 0;
5899 xmlGenericError(xmlGenericErrorContext,
5900 "HPP: entering CONTENT\n");
5903 case XML_PARSER_CDATA_SECTION:
5904 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
5905 "HPP: internal error, state == CDATA\n",
5907 ctxt->instate = XML_PARSER_CONTENT;
5908 ctxt->checkIndex = 0;
5910 xmlGenericError(xmlGenericErrorContext,
5911 "HPP: entering CONTENT\n");
5914 case XML_PARSER_DTD:
5915 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
5916 "HPP: internal error, state == DTD\n",
5918 ctxt->instate = XML_PARSER_CONTENT;
5919 ctxt->checkIndex = 0;
5921 xmlGenericError(xmlGenericErrorContext,
5922 "HPP: entering CONTENT\n");
5925 case XML_PARSER_COMMENT:
5926 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
5927 "HPP: internal error, state == COMMENT\n",
5929 ctxt->instate = XML_PARSER_CONTENT;
5930 ctxt->checkIndex = 0;
5932 xmlGenericError(xmlGenericErrorContext,
5933 "HPP: entering CONTENT\n");
5937 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
5938 "HPP: internal error, state == PI\n",
5940 ctxt->instate = XML_PARSER_CONTENT;
5941 ctxt->checkIndex = 0;
5943 xmlGenericError(xmlGenericErrorContext,
5944 "HPP: entering CONTENT\n");
5947 case XML_PARSER_ENTITY_DECL:
5948 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
5949 "HPP: internal error, state == ENTITY_DECL\n",
5951 ctxt->instate = XML_PARSER_CONTENT;
5952 ctxt->checkIndex = 0;
5954 xmlGenericError(xmlGenericErrorContext,
5955 "HPP: entering CONTENT\n");
5958 case XML_PARSER_ENTITY_VALUE:
5959 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
5960 "HPP: internal error, state == ENTITY_VALUE\n",
5962 ctxt->instate = XML_PARSER_CONTENT;
5963 ctxt->checkIndex = 0;
5965 xmlGenericError(xmlGenericErrorContext,
5966 "HPP: entering DTD\n");
5969 case XML_PARSER_ATTRIBUTE_VALUE:
5970 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
5971 "HPP: internal error, state == ATTRIBUTE_VALUE\n",
5973 ctxt->instate = XML_PARSER_START_TAG;
5974 ctxt->checkIndex = 0;
5976 xmlGenericError(xmlGenericErrorContext,
5977 "HPP: entering START_TAG\n");
5980 case XML_PARSER_SYSTEM_LITERAL:
5981 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
5982 "HPP: internal error, state == XML_PARSER_SYSTEM_LITERAL\n",
5984 ctxt->instate = XML_PARSER_CONTENT;
5985 ctxt->checkIndex = 0;
5987 xmlGenericError(xmlGenericErrorContext,
5988 "HPP: entering CONTENT\n");
5991 case XML_PARSER_IGNORE:
5992 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
5993 "HPP: internal error, state == XML_PARSER_IGNORE\n",
5995 ctxt->instate = XML_PARSER_CONTENT;
5996 ctxt->checkIndex = 0;
5998 xmlGenericError(xmlGenericErrorContext,
5999 "HPP: entering CONTENT\n");
6002 case XML_PARSER_PUBLIC_LITERAL:
6003 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
6004 "HPP: internal error, state == XML_PARSER_LITERAL\n",
6006 ctxt->instate = XML_PARSER_CONTENT;
6007 ctxt->checkIndex = 0;
6009 xmlGenericError(xmlGenericErrorContext,
6010 "HPP: entering CONTENT\n");
6017 if ((avail == 0) && (terminate)) {
6018 htmlAutoCloseOnEnd(ctxt);
6019 if ((ctxt->nameNr == 0) && (ctxt->instate != XML_PARSER_EOF)) {
6021 * SAX: end of the document processing.
6023 ctxt->instate = XML_PARSER_EOF;
6024 if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
6025 ctxt->sax->endDocument(ctxt->userData);
6028 if ((!(ctxt->options & HTML_PARSE_NODEFDTD)) && (ctxt->myDoc != NULL) &&
6029 ((terminate) || (ctxt->instate == XML_PARSER_EOF) ||
6030 (ctxt->instate == XML_PARSER_EPILOG))) {
6032 dtd = xmlGetIntSubset(ctxt->myDoc);
6034 ctxt->myDoc->intSubset =
6035 xmlCreateIntSubset(ctxt->myDoc, BAD_CAST "html",
6036 BAD_CAST "-//W3C//DTD HTML 4.0 Transitional//EN",
6037 BAD_CAST "http://www.w3.org/TR/REC-html40/loose.dtd");
6040 xmlGenericError(xmlGenericErrorContext, "HPP: done %d\n", ret);
6047 * @ctxt: an HTML parser context
6048 * @chunk: an char array
6049 * @size: the size in byte of the chunk
6050 * @terminate: last chunk indicator
6052 * Parse a Chunk of memory
6054 * Returns zero if no error, the xmlParserErrors otherwise.
6057 htmlParseChunk(htmlParserCtxtPtr ctxt, const char *chunk, int size,
6059 if ((ctxt == NULL) || (ctxt->input == NULL)) {
6060 htmlParseErr(ctxt, XML_ERR_INTERNAL_ERROR,
6061 "htmlParseChunk: context error\n", NULL, NULL);
6062 return(XML_ERR_INTERNAL_ERROR);
6064 if ((size > 0) && (chunk != NULL) && (ctxt->input != NULL) &&
6065 (ctxt->input->buf != NULL) && (ctxt->instate != XML_PARSER_EOF)) {
6066 size_t base = xmlBufGetInputBase(ctxt->input->buf->buffer, ctxt->input);
6067 size_t cur = ctxt->input->cur - ctxt->input->base;
6070 res = xmlParserInputBufferPush(ctxt->input->buf, size, chunk);
6072 ctxt->errNo = XML_PARSER_EOF;
6073 ctxt->disableSAX = 1;
6074 return (XML_PARSER_EOF);
6076 xmlBufSetInputBaseCur(ctxt->input->buf->buffer, ctxt->input, base, cur);
6078 xmlGenericError(xmlGenericErrorContext, "HPP: pushed %d\n", size);
6082 if ((terminate) || (ctxt->input->buf->buffer->use > 80))
6083 htmlParseTryOrFinish(ctxt, terminate);
6085 } else if (ctxt->instate != XML_PARSER_EOF) {
6086 if ((ctxt->input != NULL) && ctxt->input->buf != NULL) {
6087 xmlParserInputBufferPtr in = ctxt->input->buf;
6088 if ((in->encoder != NULL) && (in->buffer != NULL) &&
6089 (in->raw != NULL)) {
6091 size_t base = xmlBufGetInputBase(in->buffer, ctxt->input);
6092 size_t current = ctxt->input->cur - ctxt->input->base;
6094 nbchars = xmlCharEncInput(in, terminate);
6096 htmlParseErr(ctxt, XML_ERR_INVALID_ENCODING,
6097 "encoder error\n", NULL, NULL);
6098 return(XML_ERR_INVALID_ENCODING);
6100 xmlBufSetInputBaseCur(in->buffer, ctxt->input, base, current);
6104 htmlParseTryOrFinish(ctxt, terminate);
6106 if ((ctxt->instate != XML_PARSER_EOF) &&
6107 (ctxt->instate != XML_PARSER_EPILOG) &&
6108 (ctxt->instate != XML_PARSER_MISC)) {
6109 ctxt->errNo = XML_ERR_DOCUMENT_END;
6110 ctxt->wellFormed = 0;
6112 if (ctxt->instate != XML_PARSER_EOF) {
6113 if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
6114 ctxt->sax->endDocument(ctxt->userData);
6116 ctxt->instate = XML_PARSER_EOF;
6118 return((xmlParserErrors) ctxt->errNo);
6121 /************************************************************************
6123 * User entry points *
6125 ************************************************************************/
6128 * htmlCreatePushParserCtxt:
6129 * @sax: a SAX handler
6130 * @user_data: The user data returned on SAX callbacks
6131 * @chunk: a pointer to an array of chars
6132 * @size: number of chars in the array
6133 * @filename: an optional file name or URI
6134 * @enc: an optional encoding
6136 * Create a parser context for using the HTML parser in push mode
6137 * The value of @filename is used for fetching external entities
6138 * and error/warning reports.
6140 * Returns the new parser context or NULL
6143 htmlCreatePushParserCtxt(htmlSAXHandlerPtr sax, void *user_data,
6144 const char *chunk, int size, const char *filename,
6145 xmlCharEncoding enc) {
6146 htmlParserCtxtPtr ctxt;
6147 htmlParserInputPtr inputStream;
6148 xmlParserInputBufferPtr buf;
6152 buf = xmlAllocParserInputBuffer(enc);
6153 if (buf == NULL) return(NULL);
6155 ctxt = htmlNewParserCtxt();
6157 xmlFreeParserInputBuffer(buf);
6160 if(enc==XML_CHAR_ENCODING_UTF8 || buf->encoder)
6161 ctxt->charset=XML_CHAR_ENCODING_UTF8;
6163 if (ctxt->sax != (xmlSAXHandlerPtr) &htmlDefaultSAXHandler)
6165 ctxt->sax = (htmlSAXHandlerPtr) xmlMalloc(sizeof(htmlSAXHandler));
6166 if (ctxt->sax == NULL) {
6171 memcpy(ctxt->sax, sax, sizeof(htmlSAXHandler));
6172 if (user_data != NULL)
6173 ctxt->userData = user_data;
6175 if (filename == NULL) {
6176 ctxt->directory = NULL;
6178 ctxt->directory = xmlParserGetDirectory(filename);
6181 inputStream = htmlNewInputStream(ctxt);
6182 if (inputStream == NULL) {
6183 xmlFreeParserCtxt(ctxt);
6188 if (filename == NULL)
6189 inputStream->filename = NULL;
6191 inputStream->filename = (char *)
6192 xmlCanonicPath((const xmlChar *) filename);
6193 inputStream->buf = buf;
6194 xmlBufResetInput(buf->buffer, inputStream);
6196 inputPush(ctxt, inputStream);
6198 if ((size > 0) && (chunk != NULL) && (ctxt->input != NULL) &&
6199 (ctxt->input->buf != NULL)) {
6200 size_t base = xmlBufGetInputBase(ctxt->input->buf->buffer, ctxt->input);
6201 size_t cur = ctxt->input->cur - ctxt->input->base;
6203 xmlParserInputBufferPush(ctxt->input->buf, size, chunk);
6205 xmlBufSetInputBaseCur(ctxt->input->buf->buffer, ctxt->input, base, cur);
6207 xmlGenericError(xmlGenericErrorContext, "HPP: pushed %d\n", size);
6210 ctxt->progressive = 1;
6214 #endif /* LIBXML_PUSH_ENABLED */
6218 * @cur: a pointer to an array of xmlChar
6219 * @encoding: a free form C string describing the HTML document encoding, or NULL
6220 * @sax: the SAX handler block
6221 * @userData: if using SAX, this pointer will be provided on callbacks.
6223 * Parse an HTML in-memory document. If sax is not NULL, use the SAX callbacks
6224 * to handle parse events. If sax is NULL, fallback to the default DOM
6225 * behavior and return a tree.
6227 * Returns the resulting document tree unless SAX is NULL or the document is
6232 htmlSAXParseDoc(xmlChar *cur, const char *encoding, htmlSAXHandlerPtr sax, void *userData) {
6234 htmlParserCtxtPtr ctxt;
6238 if (cur == NULL) return(NULL);
6241 ctxt = htmlCreateDocParserCtxt(cur, encoding);
6242 if (ctxt == NULL) return(NULL);
6244 if (ctxt->sax != NULL) xmlFree (ctxt->sax);
6246 ctxt->userData = userData;
6249 htmlParseDocument(ctxt);
6253 ctxt->userData = NULL;
6255 htmlFreeParserCtxt(ctxt);
6262 * @cur: a pointer to an array of xmlChar
6263 * @encoding: a free form C string describing the HTML document encoding, or NULL
6265 * parse an HTML in-memory document and build a tree.
6267 * Returns the resulting document tree
6271 htmlParseDoc(xmlChar *cur, const char *encoding) {
6272 return(htmlSAXParseDoc(cur, encoding, NULL, NULL));
6277 * htmlCreateFileParserCtxt:
6278 * @filename: the filename
6279 * @encoding: a free form C string describing the HTML document encoding, or NULL
6281 * Create a parser context for a file content.
6282 * Automatic support for ZLIB/Compress compressed document is provided
6283 * by default if found at compile-time.
6285 * Returns the new parser context or NULL
6288 htmlCreateFileParserCtxt(const char *filename, const char *encoding)
6290 htmlParserCtxtPtr ctxt;
6291 htmlParserInputPtr inputStream;
6292 char *canonicFilename;
6293 /* htmlCharEncoding enc; */
6294 xmlChar *content, *content_line = (xmlChar *) "charset=";
6296 if (filename == NULL)
6299 ctxt = htmlNewParserCtxt();
6303 canonicFilename = (char *) xmlCanonicPath((const xmlChar *) filename);
6304 if (canonicFilename == NULL) {
6305 #ifdef LIBXML_SAX1_ENABLED
6306 if (xmlDefaultSAXHandler.error != NULL) {
6307 xmlDefaultSAXHandler.error(NULL, "out of memory\n");
6310 xmlFreeParserCtxt(ctxt);
6314 inputStream = xmlLoadExternalEntity(canonicFilename, NULL, ctxt);
6315 xmlFree(canonicFilename);
6316 if (inputStream == NULL) {
6317 xmlFreeParserCtxt(ctxt);
6321 inputPush(ctxt, inputStream);
6325 size_t l = strlen(encoding);
6328 content = xmlMallocAtomic (xmlStrlen(content_line) + l + 1);
6330 strcpy ((char *)content, (char *)content_line);
6331 strcat ((char *)content, (char *)encoding);
6332 htmlCheckEncoding (ctxt, content);
6343 * @filename: the filename
6344 * @encoding: a free form C string describing the HTML document encoding, or NULL
6345 * @sax: the SAX handler block
6346 * @userData: if using SAX, this pointer will be provided on callbacks.
6348 * parse an HTML file and build a tree. Automatic support for ZLIB/Compress
6349 * compressed document is provided by default if found at compile-time.
6350 * It use the given SAX function block to handle the parsing callback.
6351 * If sax is NULL, fallback to the default DOM tree building routines.
6353 * Returns the resulting document tree unless SAX is NULL or the document is
6358 htmlSAXParseFile(const char *filename, const char *encoding, htmlSAXHandlerPtr sax,
6361 htmlParserCtxtPtr ctxt;
6362 htmlSAXHandlerPtr oldsax = NULL;
6366 ctxt = htmlCreateFileParserCtxt(filename, encoding);
6367 if (ctxt == NULL) return(NULL);
6371 ctxt->userData = userData;
6374 htmlParseDocument(ctxt);
6379 ctxt->userData = NULL;
6381 htmlFreeParserCtxt(ctxt);
6388 * @filename: the filename
6389 * @encoding: a free form C string describing the HTML document encoding, or NULL
6391 * parse an HTML file and build a tree. Automatic support for ZLIB/Compress
6392 * compressed document is provided by default if found at compile-time.
6394 * Returns the resulting document tree
6398 htmlParseFile(const char *filename, const char *encoding) {
6399 return(htmlSAXParseFile(filename, encoding, NULL, NULL));
6403 * htmlHandleOmittedElem:
6406 * Set and return the previous value for handling HTML omitted tags.
6408 * Returns the last value for 0 for no handling, 1 for auto insertion.
6412 htmlHandleOmittedElem(int val) {
6413 int old = htmlOmittedDefaultValue;
6415 htmlOmittedDefaultValue = val;
6420 * htmlElementAllowedHere:
6421 * @parent: HTML parent element
6422 * @elt: HTML element
6424 * Checks whether an HTML element may be a direct child of a parent element.
6425 * Note - doesn't check for deprecated elements
6427 * Returns 1 if allowed; 0 otherwise.
6430 htmlElementAllowedHere(const htmlElemDesc* parent, const xmlChar* elt) {
6433 if ( ! elt || ! parent || ! parent->subelts )
6436 for ( p = parent->subelts; *p; ++p )
6437 if ( !xmlStrcmp((const xmlChar *)*p, elt) )
6443 * htmlElementStatusHere:
6444 * @parent: HTML parent element
6445 * @elt: HTML element
6447 * Checks whether an HTML element may be a direct child of a parent element.
6448 * and if so whether it is valid or deprecated.
6450 * Returns one of HTML_VALID, HTML_DEPRECATED, HTML_INVALID
6453 htmlElementStatusHere(const htmlElemDesc* parent, const htmlElemDesc* elt) {
6454 if ( ! parent || ! elt )
6455 return HTML_INVALID ;
6456 if ( ! htmlElementAllowedHere(parent, (const xmlChar*) elt->name ) )
6457 return HTML_INVALID ;
6459 return ( elt->dtd == 0 ) ? HTML_VALID : HTML_DEPRECATED ;
6463 * @elt: HTML element
6464 * @attr: HTML attribute
6465 * @legacy: whether to allow deprecated attributes
6467 * Checks whether an attribute is valid for an element
6468 * Has full knowledge of Required and Deprecated attributes
6470 * Returns one of HTML_REQUIRED, HTML_VALID, HTML_DEPRECATED, HTML_INVALID
6473 htmlAttrAllowed(const htmlElemDesc* elt, const xmlChar* attr, int legacy) {
6476 if ( !elt || ! attr )
6477 return HTML_INVALID ;
6479 if ( elt->attrs_req )
6480 for ( p = elt->attrs_req; *p; ++p)
6481 if ( !xmlStrcmp((const xmlChar*)*p, attr) )
6482 return HTML_REQUIRED ;
6484 if ( elt->attrs_opt )
6485 for ( p = elt->attrs_opt; *p; ++p)
6486 if ( !xmlStrcmp((const xmlChar*)*p, attr) )
6489 if ( legacy && elt->attrs_depr )
6490 for ( p = elt->attrs_depr; *p; ++p)
6491 if ( !xmlStrcmp((const xmlChar*)*p, attr) )
6492 return HTML_DEPRECATED ;
6494 return HTML_INVALID ;
6498 * @node: an htmlNodePtr in a tree
6499 * @legacy: whether to allow deprecated elements (YES is faster here
6500 * for Element nodes)
6502 * Checks whether the tree node is valid. Experimental (the author
6503 * only uses the HTML enhancements in a SAX parser)
6505 * Return: for Element nodes, a return from htmlElementAllowedHere (if
6506 * legacy allowed) or htmlElementStatusHere (otherwise).
6507 * for Attribute nodes, a return from htmlAttrAllowed
6508 * for other nodes, HTML_NA (no checks performed)
6511 htmlNodeStatus(const htmlNodePtr node, int legacy) {
6513 return HTML_INVALID ;
6515 switch ( node->type ) {
6516 case XML_ELEMENT_NODE:
6518 ? ( htmlElementAllowedHere (
6519 htmlTagLookup(node->parent->name) , node->name
6520 ) ? HTML_VALID : HTML_INVALID )
6521 : htmlElementStatusHere(
6522 htmlTagLookup(node->parent->name) ,
6523 htmlTagLookup(node->name) )
6525 case XML_ATTRIBUTE_NODE:
6526 return htmlAttrAllowed(
6527 htmlTagLookup(node->parent->name) , node->name, legacy) ;
6528 default: return HTML_NA ;
6531 /************************************************************************
6533 * New set (2.6.0) of simpler and more flexible APIs *
6535 ************************************************************************/
6540 * Free a string if it is not owned by the "dict" dictionnary in the
6543 #define DICT_FREE(str) \
6544 if ((str) && ((!dict) || \
6545 (xmlDictOwns(dict, (const xmlChar *)(str)) == 0))) \
6546 xmlFree((char *)(str));
6550 * @ctxt: an HTML parser context
6552 * Reset a parser context
6555 htmlCtxtReset(htmlParserCtxtPtr ctxt)
6557 xmlParserInputPtr input;
6566 while ((input = inputPop(ctxt)) != NULL) { /* Non consuming */
6567 xmlFreeInputStream(input);
6573 if (ctxt->spaceTab != NULL) {
6574 ctxt->spaceTab[0] = -1;
6575 ctxt->space = &ctxt->spaceTab[0];
6587 DICT_FREE(ctxt->version);
6588 ctxt->version = NULL;
6589 DICT_FREE(ctxt->encoding);
6590 ctxt->encoding = NULL;
6591 DICT_FREE(ctxt->directory);
6592 ctxt->directory = NULL;
6593 DICT_FREE(ctxt->extSubURI);
6594 ctxt->extSubURI = NULL;
6595 DICT_FREE(ctxt->extSubSystem);
6596 ctxt->extSubSystem = NULL;
6597 if (ctxt->myDoc != NULL)
6598 xmlFreeDoc(ctxt->myDoc);
6601 ctxt->standalone = -1;
6602 ctxt->hasExternalSubset = 0;
6603 ctxt->hasPErefs = 0;
6606 ctxt->instate = XML_PARSER_START;
6609 ctxt->wellFormed = 1;
6610 ctxt->nsWellFormed = 1;
6611 ctxt->disableSAX = 0;
6613 ctxt->vctxt.userData = ctxt;
6614 ctxt->vctxt.error = xmlParserValidityError;
6615 ctxt->vctxt.warning = xmlParserValidityWarning;
6616 ctxt->record_info = 0;
6618 ctxt->checkIndex = 0;
6620 ctxt->errNo = XML_ERR_OK;
6622 ctxt->charset = XML_CHAR_ENCODING_NONE;
6623 ctxt->catalogs = NULL;
6624 xmlInitNodeInfoSeq(&ctxt->node_seq);
6626 if (ctxt->attsDefault != NULL) {
6627 xmlHashFree(ctxt->attsDefault, (xmlHashDeallocator) xmlFree);
6628 ctxt->attsDefault = NULL;
6630 if (ctxt->attsSpecial != NULL) {
6631 xmlHashFree(ctxt->attsSpecial, NULL);
6632 ctxt->attsSpecial = NULL;
6637 * htmlCtxtUseOptions:
6638 * @ctxt: an HTML parser context
6639 * @options: a combination of htmlParserOption(s)
6641 * Applies the options to the parser context
6643 * Returns 0 in case of success, the set of unknown or unimplemented options
6647 htmlCtxtUseOptions(htmlParserCtxtPtr ctxt, int options)
6652 if (options & HTML_PARSE_NOWARNING) {
6653 ctxt->sax->warning = NULL;
6654 ctxt->vctxt.warning = NULL;
6655 options -= XML_PARSE_NOWARNING;
6656 ctxt->options |= XML_PARSE_NOWARNING;
6658 if (options & HTML_PARSE_NOERROR) {
6659 ctxt->sax->error = NULL;
6660 ctxt->vctxt.error = NULL;
6661 ctxt->sax->fatalError = NULL;
6662 options -= XML_PARSE_NOERROR;
6663 ctxt->options |= XML_PARSE_NOERROR;
6665 if (options & HTML_PARSE_PEDANTIC) {
6667 options -= XML_PARSE_PEDANTIC;
6668 ctxt->options |= XML_PARSE_PEDANTIC;
6671 if (options & XML_PARSE_NOBLANKS) {
6672 ctxt->keepBlanks = 0;
6673 ctxt->sax->ignorableWhitespace = xmlSAX2IgnorableWhitespace;
6674 options -= XML_PARSE_NOBLANKS;
6675 ctxt->options |= XML_PARSE_NOBLANKS;
6677 ctxt->keepBlanks = 1;
6678 if (options & HTML_PARSE_RECOVER) {
6680 options -= HTML_PARSE_RECOVER;
6683 if (options & HTML_PARSE_COMPACT) {
6684 ctxt->options |= HTML_PARSE_COMPACT;
6685 options -= HTML_PARSE_COMPACT;
6687 if (options & XML_PARSE_HUGE) {
6688 ctxt->options |= XML_PARSE_HUGE;
6689 options -= XML_PARSE_HUGE;
6691 if (options & HTML_PARSE_NODEFDTD) {
6692 ctxt->options |= HTML_PARSE_NODEFDTD;
6693 options -= HTML_PARSE_NODEFDTD;
6695 if (options & HTML_PARSE_IGNORE_ENC) {
6696 ctxt->options |= HTML_PARSE_IGNORE_ENC;
6697 options -= HTML_PARSE_IGNORE_ENC;
6699 if (options & HTML_PARSE_NOIMPLIED) {
6700 ctxt->options |= HTML_PARSE_NOIMPLIED;
6701 options -= HTML_PARSE_NOIMPLIED;
6703 ctxt->dictNames = 0;
6709 * @ctxt: an HTML parser context
6710 * @URL: the base URL to use for the document
6711 * @encoding: the document encoding, or NULL
6712 * @options: a combination of htmlParserOption(s)
6713 * @reuse: keep the context for reuse
6715 * Common front-end for the htmlRead functions
6717 * Returns the resulting document tree or NULL
6720 htmlDoRead(htmlParserCtxtPtr ctxt, const char *URL, const char *encoding,
6721 int options, int reuse)
6725 htmlCtxtUseOptions(ctxt, options);
6727 if (encoding != NULL) {
6728 xmlCharEncodingHandlerPtr hdlr;
6730 hdlr = xmlFindCharEncodingHandler(encoding);
6732 xmlSwitchToEncoding(ctxt, hdlr);
6733 if (ctxt->input->encoding != NULL)
6734 xmlFree((xmlChar *) ctxt->input->encoding);
6735 ctxt->input->encoding = xmlStrdup((xmlChar *)encoding);
6738 if ((URL != NULL) && (ctxt->input != NULL) &&
6739 (ctxt->input->filename == NULL))
6740 ctxt->input->filename = (char *) xmlStrdup((const xmlChar *) URL);
6741 htmlParseDocument(ctxt);
6745 if ((ctxt->dictNames) &&
6747 (ret->dict == ctxt->dict))
6749 xmlFreeParserCtxt(ctxt);
6756 * @cur: a pointer to a zero terminated string
6757 * @URL: the base URL to use for the document
6758 * @encoding: the document encoding, or NULL
6759 * @options: a combination of htmlParserOption(s)
6761 * parse an XML in-memory document and build a tree.
6763 * Returns the resulting document tree
6766 htmlReadDoc(const xmlChar * cur, const char *URL, const char *encoding, int options)
6768 htmlParserCtxtPtr ctxt;
6774 ctxt = htmlCreateDocParserCtxt(cur, NULL);
6777 return (htmlDoRead(ctxt, URL, encoding, options, 0));
6782 * @filename: a file or URL
6783 * @encoding: the document encoding, or NULL
6784 * @options: a combination of htmlParserOption(s)
6786 * parse an XML file from the filesystem or the network.
6788 * Returns the resulting document tree
6791 htmlReadFile(const char *filename, const char *encoding, int options)
6793 htmlParserCtxtPtr ctxt;
6796 ctxt = htmlCreateFileParserCtxt(filename, encoding);
6799 return (htmlDoRead(ctxt, NULL, NULL, options, 0));
6804 * @buffer: a pointer to a char array
6805 * @size: the size of the array
6806 * @URL: the base URL to use for the document
6807 * @encoding: the document encoding, or NULL
6808 * @options: a combination of htmlParserOption(s)
6810 * parse an XML in-memory document and build a tree.
6812 * Returns the resulting document tree
6815 htmlReadMemory(const char *buffer, int size, const char *URL, const char *encoding, int options)
6817 htmlParserCtxtPtr ctxt;
6820 ctxt = xmlCreateMemoryParserCtxt(buffer, size);
6823 htmlDefaultSAXHandlerInit();
6824 if (ctxt->sax != NULL)
6825 memcpy(ctxt->sax, &htmlDefaultSAXHandler, sizeof(xmlSAXHandlerV1));
6826 return (htmlDoRead(ctxt, URL, encoding, options, 0));
6831 * @fd: an open file descriptor
6832 * @URL: the base URL to use for the document
6833 * @encoding: the document encoding, or NULL
6834 * @options: a combination of htmlParserOption(s)
6836 * parse an XML from a file descriptor and build a tree.
6838 * Returns the resulting document tree
6841 htmlReadFd(int fd, const char *URL, const char *encoding, int options)
6843 htmlParserCtxtPtr ctxt;
6844 xmlParserInputBufferPtr input;
6845 xmlParserInputPtr stream;
6852 input = xmlParserInputBufferCreateFd(fd, XML_CHAR_ENCODING_NONE);
6855 ctxt = xmlNewParserCtxt();
6857 xmlFreeParserInputBuffer(input);
6860 stream = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE);
6861 if (stream == NULL) {
6862 xmlFreeParserInputBuffer(input);
6863 xmlFreeParserCtxt(ctxt);
6866 inputPush(ctxt, stream);
6867 return (htmlDoRead(ctxt, URL, encoding, options, 0));
6872 * @ioread: an I/O read function
6873 * @ioclose: an I/O close function
6874 * @ioctx: an I/O handler
6875 * @URL: the base URL to use for the document
6876 * @encoding: the document encoding, or NULL
6877 * @options: a combination of htmlParserOption(s)
6879 * parse an HTML document from I/O functions and source and build a tree.
6881 * Returns the resulting document tree
6884 htmlReadIO(xmlInputReadCallback ioread, xmlInputCloseCallback ioclose,
6885 void *ioctx, const char *URL, const char *encoding, int options)
6887 htmlParserCtxtPtr ctxt;
6888 xmlParserInputBufferPtr input;
6889 xmlParserInputPtr stream;
6895 input = xmlParserInputBufferCreateIO(ioread, ioclose, ioctx,
6896 XML_CHAR_ENCODING_NONE);
6897 if (input == NULL) {
6898 if (ioclose != NULL)
6902 ctxt = htmlNewParserCtxt();
6904 xmlFreeParserInputBuffer(input);
6907 stream = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE);
6908 if (stream == NULL) {
6909 xmlFreeParserInputBuffer(input);
6910 xmlFreeParserCtxt(ctxt);
6913 inputPush(ctxt, stream);
6914 return (htmlDoRead(ctxt, URL, encoding, options, 0));
6919 * @ctxt: an HTML parser context
6920 * @cur: a pointer to a zero terminated string
6921 * @URL: the base URL to use for the document
6922 * @encoding: the document encoding, or NULL
6923 * @options: a combination of htmlParserOption(s)
6925 * parse an XML in-memory document and build a tree.
6926 * This reuses the existing @ctxt parser context
6928 * Returns the resulting document tree
6931 htmlCtxtReadDoc(htmlParserCtxtPtr ctxt, const xmlChar * cur,
6932 const char *URL, const char *encoding, int options)
6934 xmlParserInputPtr stream;
6942 htmlCtxtReset(ctxt);
6944 stream = xmlNewStringInputStream(ctxt, cur);
6945 if (stream == NULL) {
6948 inputPush(ctxt, stream);
6949 return (htmlDoRead(ctxt, URL, encoding, options, 1));
6954 * @ctxt: an HTML parser context
6955 * @filename: a file or URL
6956 * @encoding: the document encoding, or NULL
6957 * @options: a combination of htmlParserOption(s)
6959 * parse an XML file from the filesystem or the network.
6960 * This reuses the existing @ctxt parser context
6962 * Returns the resulting document tree
6965 htmlCtxtReadFile(htmlParserCtxtPtr ctxt, const char *filename,
6966 const char *encoding, int options)
6968 xmlParserInputPtr stream;
6970 if (filename == NULL)
6976 htmlCtxtReset(ctxt);
6978 stream = xmlLoadExternalEntity(filename, NULL, ctxt);
6979 if (stream == NULL) {
6982 inputPush(ctxt, stream);
6983 return (htmlDoRead(ctxt, NULL, encoding, options, 1));
6987 * htmlCtxtReadMemory:
6988 * @ctxt: an HTML parser context
6989 * @buffer: a pointer to a char array
6990 * @size: the size of the array
6991 * @URL: the base URL to use for the document
6992 * @encoding: the document encoding, or NULL
6993 * @options: a combination of htmlParserOption(s)
6995 * parse an XML in-memory document and build a tree.
6996 * This reuses the existing @ctxt parser context
6998 * Returns the resulting document tree
7001 htmlCtxtReadMemory(htmlParserCtxtPtr ctxt, const char *buffer, int size,
7002 const char *URL, const char *encoding, int options)
7004 xmlParserInputBufferPtr input;
7005 xmlParserInputPtr stream;
7013 htmlCtxtReset(ctxt);
7015 input = xmlParserInputBufferCreateMem(buffer, size, XML_CHAR_ENCODING_NONE);
7016 if (input == NULL) {
7020 stream = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE);
7021 if (stream == NULL) {
7022 xmlFreeParserInputBuffer(input);
7026 inputPush(ctxt, stream);
7027 return (htmlDoRead(ctxt, URL, encoding, options, 1));
7032 * @ctxt: an HTML parser context
7033 * @fd: an open file descriptor
7034 * @URL: the base URL to use for the document
7035 * @encoding: the document encoding, or NULL
7036 * @options: a combination of htmlParserOption(s)
7038 * parse an XML from a file descriptor and build a tree.
7039 * This reuses the existing @ctxt parser context
7041 * Returns the resulting document tree
7044 htmlCtxtReadFd(htmlParserCtxtPtr ctxt, int fd,
7045 const char *URL, const char *encoding, int options)
7047 xmlParserInputBufferPtr input;
7048 xmlParserInputPtr stream;
7056 htmlCtxtReset(ctxt);
7059 input = xmlParserInputBufferCreateFd(fd, XML_CHAR_ENCODING_NONE);
7062 stream = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE);
7063 if (stream == NULL) {
7064 xmlFreeParserInputBuffer(input);
7067 inputPush(ctxt, stream);
7068 return (htmlDoRead(ctxt, URL, encoding, options, 1));
7073 * @ctxt: an HTML parser context
7074 * @ioread: an I/O read function
7075 * @ioclose: an I/O close function
7076 * @ioctx: an I/O handler
7077 * @URL: the base URL to use for the document
7078 * @encoding: the document encoding, or NULL
7079 * @options: a combination of htmlParserOption(s)
7081 * parse an HTML document from I/O functions and source and build a tree.
7082 * This reuses the existing @ctxt parser context
7084 * Returns the resulting document tree
7087 htmlCtxtReadIO(htmlParserCtxtPtr ctxt, xmlInputReadCallback ioread,
7088 xmlInputCloseCallback ioclose, void *ioctx,
7090 const char *encoding, int options)
7092 xmlParserInputBufferPtr input;
7093 xmlParserInputPtr stream;
7101 htmlCtxtReset(ctxt);
7103 input = xmlParserInputBufferCreateIO(ioread, ioclose, ioctx,
7104 XML_CHAR_ENCODING_NONE);
7105 if (input == NULL) {
7106 if (ioclose != NULL)
7110 stream = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE);
7111 if (stream == NULL) {
7112 xmlFreeParserInputBuffer(input);
7115 inputPush(ctxt, stream);
7116 return (htmlDoRead(ctxt, URL, encoding, options, 1));
7119 #define bottom_HTMLparser
7120 #include "elfgcchack.h"
7121 #endif /* LIBXML_HTML_ENABLED */