2 * HTMLparser.c : an HTML 4.0 non-verifying parser
4 * See Copyright for the status of this software.
11 #ifdef LIBXML_HTML_ENABLED
17 #include <libxml/xmlmemory.h>
18 #include <libxml/tree.h>
19 #include <libxml/parser.h>
20 #include <libxml/parserInternals.h>
21 #include <libxml/xmlerror.h>
22 #include <libxml/HTMLparser.h>
23 #include <libxml/HTMLtree.h>
24 #include <libxml/entities.h>
25 #include <libxml/encoding.h>
26 #include <libxml/valid.h>
27 #include <libxml/xmlIO.h>
28 #include <libxml/globals.h>
29 #include <libxml/uri.h>
34 #define HTML_MAX_NAMELEN 1000
35 #define HTML_PARSER_BIG_BUFFER_SIZE 1000
36 #define HTML_PARSER_BUFFER_SIZE 100
39 /* #define DEBUG_PUSH */
41 static int htmlOmittedDefaultValue
= 1;
43 xmlChar
* htmlDecodeEntities(htmlParserCtxtPtr ctxt
, int len
,
44 xmlChar end
, xmlChar end2
, xmlChar end3
);
45 static void htmlParseComment(htmlParserCtxtPtr ctxt
);
47 /************************************************************************
49 * Some factorized error routines *
51 ************************************************************************/
55 * @ctxt: an HTML parser context
56 * @extra: extra information
58 * Handle a redefinition of attribute error
61 htmlErrMemory(xmlParserCtxtPtr ctxt
, const char *extra
)
63 if ((ctxt
!= NULL
) && (ctxt
->disableSAX
!= 0) &&
64 (ctxt
->instate
== XML_PARSER_EOF
))
67 ctxt
->errNo
= XML_ERR_NO_MEMORY
;
68 ctxt
->instate
= XML_PARSER_EOF
;
72 __xmlRaiseError(NULL
, NULL
, NULL
, ctxt
, NULL
, XML_FROM_PARSER
,
73 XML_ERR_NO_MEMORY
, XML_ERR_FATAL
, NULL
, 0, extra
,
75 "Memory allocation failed : %s\n", extra
);
77 __xmlRaiseError(NULL
, NULL
, NULL
, ctxt
, NULL
, XML_FROM_PARSER
,
78 XML_ERR_NO_MEMORY
, XML_ERR_FATAL
, NULL
, 0, NULL
,
79 NULL
, NULL
, 0, 0, "Memory allocation failed\n");
84 * @ctxt: an HTML parser context
85 * @error: the error number
86 * @msg: the error message
90 * Handle a fatal parser error, i.e. violating Well-Formedness constraints
92 static void LIBXML_ATTR_FORMAT(3,0)
93 htmlParseErr(xmlParserCtxtPtr ctxt
, xmlParserErrors error
,
94 const char *msg
, const xmlChar
*str1
, const xmlChar
*str2
)
96 if ((ctxt
!= NULL
) && (ctxt
->disableSAX
!= 0) &&
97 (ctxt
->instate
== XML_PARSER_EOF
))
101 __xmlRaiseError(NULL
, NULL
, NULL
, ctxt
, NULL
, XML_FROM_HTML
, error
,
102 XML_ERR_ERROR
, NULL
, 0,
103 (const char *) str1
, (const char *) str2
,
107 ctxt
->wellFormed
= 0;
112 * @ctxt: an HTML parser context
113 * @error: the error number
114 * @msg: the error message
117 * Handle a fatal parser error, i.e. violating Well-Formedness constraints
119 static void LIBXML_ATTR_FORMAT(3,0)
120 htmlParseErrInt(xmlParserCtxtPtr ctxt
, xmlParserErrors error
,
121 const char *msg
, int val
)
123 if ((ctxt
!= NULL
) && (ctxt
->disableSAX
!= 0) &&
124 (ctxt
->instate
== XML_PARSER_EOF
))
128 __xmlRaiseError(NULL
, NULL
, NULL
, ctxt
, NULL
, XML_FROM_HTML
, error
,
129 XML_ERR_ERROR
, NULL
, 0, NULL
, NULL
,
130 NULL
, val
, 0, msg
, val
);
132 ctxt
->wellFormed
= 0;
135 /************************************************************************
137 * Parser stacks related functions and macros *
139 ************************************************************************/
143 * @ctxt: an HTML parser context
144 * @value: the element name
146 * Pushes a new element name on top of the name stack
148 * Returns 0 in case of error, the index in the stack otherwise
151 htmlnamePush(htmlParserCtxtPtr ctxt
, const xmlChar
* value
)
153 if ((ctxt
->html
< 3) && (xmlStrEqual(value
, BAD_CAST
"head")))
155 if ((ctxt
->html
< 10) && (xmlStrEqual(value
, BAD_CAST
"body")))
157 if (ctxt
->nameNr
>= ctxt
->nameMax
) {
159 ctxt
->nameTab
= (const xmlChar
* *)
160 xmlRealloc((xmlChar
* *)ctxt
->nameTab
,
162 sizeof(ctxt
->nameTab
[0]));
163 if (ctxt
->nameTab
== NULL
) {
164 htmlErrMemory(ctxt
, NULL
);
168 ctxt
->nameTab
[ctxt
->nameNr
] = value
;
170 return (ctxt
->nameNr
++);
174 * @ctxt: an HTML parser context
176 * Pops the top element name from the name stack
178 * Returns the name just removed
180 static const xmlChar
*
181 htmlnamePop(htmlParserCtxtPtr ctxt
)
185 if (ctxt
->nameNr
<= 0)
188 if (ctxt
->nameNr
< 0)
190 if (ctxt
->nameNr
> 0)
191 ctxt
->name
= ctxt
->nameTab
[ctxt
->nameNr
- 1];
194 ret
= ctxt
->nameTab
[ctxt
->nameNr
];
195 ctxt
->nameTab
[ctxt
->nameNr
] = NULL
;
201 * @ctxt: an HTML parser context
202 * @value: the node info
204 * Pushes a new element name on top of the node info stack
206 * Returns 0 in case of error, the index in the stack otherwise
209 htmlNodeInfoPush(htmlParserCtxtPtr ctxt
, htmlParserNodeInfo
*value
)
211 if (ctxt
->nodeInfoNr
>= ctxt
->nodeInfoMax
) {
212 if (ctxt
->nodeInfoMax
== 0)
213 ctxt
->nodeInfoMax
= 5;
214 ctxt
->nodeInfoMax
*= 2;
215 ctxt
->nodeInfoTab
= (htmlParserNodeInfo
*)
216 xmlRealloc((htmlParserNodeInfo
*)ctxt
->nodeInfoTab
,
218 sizeof(ctxt
->nodeInfoTab
[0]));
219 if (ctxt
->nodeInfoTab
== NULL
) {
220 htmlErrMemory(ctxt
, NULL
);
224 ctxt
->nodeInfoTab
[ctxt
->nodeInfoNr
] = *value
;
225 ctxt
->nodeInfo
= &ctxt
->nodeInfoTab
[ctxt
->nodeInfoNr
];
226 return (ctxt
->nodeInfoNr
++);
231 * @ctxt: an HTML parser context
233 * Pops the top element name from the node info stack
235 * Returns 0 in case of error, the pointer to NodeInfo otherwise
237 static htmlParserNodeInfo
*
238 htmlNodeInfoPop(htmlParserCtxtPtr ctxt
)
240 if (ctxt
->nodeInfoNr
<= 0)
243 if (ctxt
->nodeInfoNr
< 0)
245 if (ctxt
->nodeInfoNr
> 0)
246 ctxt
->nodeInfo
= &ctxt
->nodeInfoTab
[ctxt
->nodeInfoNr
- 1];
248 ctxt
->nodeInfo
= NULL
;
249 return &ctxt
->nodeInfoTab
[ctxt
->nodeInfoNr
];
253 * Macros for accessing the content. Those should be used only by the parser,
256 * Dirty macros, i.e. one need to make assumption on the context to use them
258 * CUR_PTR return the current pointer to the xmlChar to be parsed.
259 * CUR returns the current xmlChar value, i.e. a 8 bit value if compiled
260 * in ISO-Latin or UTF-8, and the current 16 bit value if compiled
261 * in UNICODE mode. This should be used internally by the parser
262 * only to compare to ASCII values otherwise it would break when
263 * running with UTF-8 encoding.
264 * NXT(n) returns the n'th next xmlChar. Same as CUR is should be used only
265 * to compare on ASCII based substring.
266 * UPP(n) returns the n'th next xmlChar converted to uppercase. Same as CUR
267 * it should be used only to compare on ASCII based substring.
268 * SKIP(n) Skip n xmlChar, and must also be used only to skip ASCII defined
269 * strings without newlines within the parser.
271 * Clean macros, not dependent of an ASCII context, expect UTF-8 encoding
273 * CURRENT Returns the current char value, with the full decoding of
274 * UTF-8 if we are using this mode. It returns an int.
275 * NEXT Skip to the next character, this does the proper decoding
276 * in UTF-8 mode. It also pop-up unfinished entities on the fly.
277 * NEXTL(l) Skip the current unicode character of l xmlChars long.
278 * COPY(to) copy one char to *to, increment CUR_PTR and to accordingly
281 #define UPPER (toupper(*ctxt->input->cur))
283 #define SKIP(val) ctxt->input->cur += (val),ctxt->input->col+=(val)
285 #define NXT(val) ctxt->input->cur[(val)]
287 #define UPP(val) (toupper(ctxt->input->cur[(val)]))
289 #define CUR_PTR ctxt->input->cur
290 #define BASE_PTR ctxt->input->base
292 #define SHRINK if ((ctxt->input->cur - ctxt->input->base > 2 * INPUT_CHUNK) && \
293 (ctxt->input->end - ctxt->input->cur < 2 * INPUT_CHUNK)) \
294 xmlParserInputShrink(ctxt->input)
296 #define GROW if ((ctxt->progressive == 0) && \
297 (ctxt->input->end - ctxt->input->cur < INPUT_CHUNK)) \
298 xmlParserInputGrow(ctxt->input, INPUT_CHUNK)
300 #define CURRENT ((int) (*ctxt->input->cur))
302 #define SKIP_BLANKS htmlSkipBlankChars(ctxt)
304 /* Imported from XML */
306 /* #define CUR (ctxt->token ? ctxt->token : (int) (*ctxt->input->cur)) */
307 #define CUR ((int) (*ctxt->input->cur))
308 #define NEXT xmlNextChar(ctxt)
310 #define RAW (ctxt->token ? -1 : (*ctxt->input->cur))
313 #define NEXTL(l) do { \
314 if (*(ctxt->input->cur) == '\n') { \
315 ctxt->input->line++; ctxt->input->col = 1; \
316 } else ctxt->input->col++; \
317 ctxt->token = 0; ctxt->input->cur += l; \
322 if (*ctxt->input->cur == '%') xmlParserHandlePEReference(ctxt); \
323 if (*ctxt->input->cur == '&') xmlParserHandleReference(ctxt);
326 #define CUR_CHAR(l) htmlCurrentChar(ctxt, &l)
327 #define CUR_SCHAR(s, l) xmlStringCurrentChar(ctxt, s, &l)
329 #define COPY_BUF(l,b,i,v) \
330 if (l == 1) b[i++] = (xmlChar) v; \
331 else i += xmlCopyChar(l,&b[i],v)
335 * @the HTML parser context
337 * Ty to find and encoding in the current data available in the input
338 * buffer this is needed to try to switch to the proper encoding when
339 * one face a character error.
340 * That's an heuristic, since it's operating outside of parsing it could
341 * try to use a meta which had been commented out, that's the reason it
342 * should only be used in case of error, not as a default.
344 * Returns an encoding string or NULL if not found, the string need to
348 htmlFindEncoding(xmlParserCtxtPtr ctxt
) {
349 const xmlChar
*start
, *cur
, *end
;
351 if ((ctxt
== NULL
) || (ctxt
->input
== NULL
) ||
352 (ctxt
->input
->encoding
!= NULL
) || (ctxt
->input
->buf
== NULL
) ||
353 (ctxt
->input
->buf
->encoder
!= NULL
))
355 if ((ctxt
->input
->cur
== NULL
) || (ctxt
->input
->end
== NULL
))
358 start
= ctxt
->input
->cur
;
359 end
= ctxt
->input
->end
;
360 /* we also expect the input buffer to be zero terminated */
364 cur
= xmlStrcasestr(start
, BAD_CAST
"HTTP-EQUIV");
367 cur
= xmlStrcasestr(cur
, BAD_CAST
"CONTENT");
370 cur
= xmlStrcasestr(cur
, BAD_CAST
"CHARSET=");
375 while (((*cur
>= 'A') && (*cur
<= 'Z')) ||
376 ((*cur
>= 'a') && (*cur
<= 'z')) ||
377 ((*cur
>= '0') && (*cur
<= '9')) ||
378 (*cur
== '-') || (*cur
== '_') || (*cur
== ':') || (*cur
== '/'))
382 return(xmlStrndup(start
, cur
- start
));
387 * @ctxt: the HTML parser context
388 * @len: pointer to the length of the char read
390 * The current char value, if using UTF-8 this may actually span multiple
391 * bytes in the input buffer. Implement the end of line normalization:
392 * 2.11 End-of-Line Handling
393 * If the encoding is unspecified, in the case we find an ISO-Latin-1
394 * char, then the encoding converter is plugged in automatically.
396 * Returns the current char value and its length
400 htmlCurrentChar(xmlParserCtxtPtr ctxt
, int *len
) {
401 const unsigned char *cur
;
405 if (ctxt
->instate
== XML_PARSER_EOF
)
408 if (ctxt
->token
!= 0) {
412 if (ctxt
->charset
!= XML_CHAR_ENCODING_UTF8
) {
414 xmlCharEncodingHandlerPtr handler
;
417 * Assume it's a fixed length encoding (1) with
418 * a compatible encoding for the ASCII set, since
419 * HTML constructs only use < 128 chars
421 if ((int) *ctxt
->input
->cur
< 0x80) {
423 if ((*ctxt
->input
->cur
== 0) &&
424 (ctxt
->input
->cur
< ctxt
->input
->end
)) {
425 htmlParseErrInt(ctxt
, XML_ERR_INVALID_CHAR
,
426 "Char 0x%X out of allowed range\n", 0);
429 return((int) *ctxt
->input
->cur
);
433 * Humm this is bad, do an automatic flow conversion
435 guess
= htmlFindEncoding(ctxt
);
437 xmlSwitchEncoding(ctxt
, XML_CHAR_ENCODING_8859_1
);
439 if (ctxt
->input
->encoding
!= NULL
)
440 xmlFree((xmlChar
*) ctxt
->input
->encoding
);
441 ctxt
->input
->encoding
= guess
;
442 handler
= xmlFindCharEncodingHandler((const char *) guess
);
443 if (handler
!= NULL
) {
445 * Don't use UTF-8 encoder which isn't required and
446 * can produce invalid UTF-8.
448 if (!xmlStrEqual(BAD_CAST handler
->name
, BAD_CAST
"UTF-8"))
449 xmlSwitchToEncoding(ctxt
, handler
);
451 htmlParseErr(ctxt
, XML_ERR_INVALID_ENCODING
,
452 "Unsupported encoding %s", guess
, NULL
);
455 ctxt
->charset
= XML_CHAR_ENCODING_UTF8
;
459 * We are supposed to handle UTF8, check it's valid
460 * From rfc2044: encoding of the Unicode values on UTF-8:
462 * UCS-4 range (hex.) UTF-8 octet sequence (binary)
463 * 0000 0000-0000 007F 0xxxxxxx
464 * 0000 0080-0000 07FF 110xxxxx 10xxxxxx
465 * 0000 0800-0000 FFFF 1110xxxx 10xxxxxx 10xxxxxx
467 * Check for the 0x110000 limit too
469 cur
= ctxt
->input
->cur
;
475 xmlParserInputGrow(ctxt
->input
, INPUT_CHUNK
);
476 cur
= ctxt
->input
->cur
;
478 if ((cur
[1] & 0xc0) != 0x80)
480 if ((c
& 0xe0) == 0xe0) {
483 xmlParserInputGrow(ctxt
->input
, INPUT_CHUNK
);
484 cur
= ctxt
->input
->cur
;
486 if ((cur
[2] & 0xc0) != 0x80)
488 if ((c
& 0xf0) == 0xf0) {
490 xmlParserInputGrow(ctxt
->input
, INPUT_CHUNK
);
491 cur
= ctxt
->input
->cur
;
493 if (((c
& 0xf8) != 0xf0) ||
494 ((cur
[3] & 0xc0) != 0x80))
498 val
= (cur
[0] & 0x7) << 18;
499 val
|= (cur
[1] & 0x3f) << 12;
500 val
|= (cur
[2] & 0x3f) << 6;
501 val
|= cur
[3] & 0x3f;
507 val
= (cur
[0] & 0xf) << 12;
508 val
|= (cur
[1] & 0x3f) << 6;
509 val
|= cur
[2] & 0x3f;
516 val
= (cur
[0] & 0x1f) << 6;
517 val
|= cur
[1] & 0x3f;
522 htmlParseErrInt(ctxt
, XML_ERR_INVALID_CHAR
,
523 "Char 0x%X out of allowed range\n", val
);
527 if ((*ctxt
->input
->cur
== 0) &&
528 (ctxt
->input
->cur
< ctxt
->input
->end
)) {
529 htmlParseErrInt(ctxt
, XML_ERR_INVALID_CHAR
,
530 "Char 0x%X out of allowed range\n", 0);
536 return((int) *ctxt
->input
->cur
);
541 * If we detect an UTF8 error that probably mean that the
542 * input encoding didn't get properly advertised in the
543 * declaration header. Report the error and switch the encoding
544 * to ISO-Latin-1 (if you don't like this policy, just declare the
550 if (ctxt
->input
->end
- ctxt
->input
->cur
>= 4) {
551 snprintf(buffer
, 149, "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n",
552 ctxt
->input
->cur
[0], ctxt
->input
->cur
[1],
553 ctxt
->input
->cur
[2], ctxt
->input
->cur
[3]);
555 snprintf(buffer
, 149, "Bytes: 0x%02X\n", ctxt
->input
->cur
[0]);
557 htmlParseErr(ctxt
, XML_ERR_INVALID_ENCODING
,
558 "Input is not proper UTF-8, indicate encoding !\n",
559 BAD_CAST buffer
, NULL
);
563 * Don't switch encodings twice. Note that if there's an encoder, we
564 * shouldn't receive invalid UTF-8 anyway.
566 * Note that if ctxt->input->buf == NULL, switching encodings is
567 * impossible, see Gitlab issue #34.
569 if ((ctxt
->input
->buf
!= NULL
) &&
570 (ctxt
->input
->buf
->encoder
== NULL
))
571 xmlSwitchEncoding(ctxt
, XML_CHAR_ENCODING_8859_1
);
573 return((int) *ctxt
->input
->cur
);
577 * htmlSkipBlankChars:
578 * @ctxt: the HTML parser context
580 * skip all blanks character found at that point in the input streams.
582 * Returns the number of space chars skipped
586 htmlSkipBlankChars(xmlParserCtxtPtr ctxt
) {
589 while (IS_BLANK_CH(*(ctxt
->input
->cur
))) {
590 if ((*ctxt
->input
->cur
== 0) &&
591 (xmlParserInputGrow(ctxt
->input
, INPUT_CHUNK
) <= 0)) {
594 if (*(ctxt
->input
->cur
) == '\n') {
595 ctxt
->input
->line
++; ctxt
->input
->col
= 1;
596 } else ctxt
->input
->col
++;
598 if (*ctxt
->input
->cur
== 0)
599 xmlParserInputGrow(ctxt
->input
, INPUT_CHUNK
);
609 /************************************************************************
611 * The list of HTML elements and their properties *
613 ************************************************************************/
616 * Start Tag: 1 means the start tag can be omitted
617 * End Tag: 1 means the end tag can be omitted
618 * 2 means it's forbidden (empty elements)
619 * 3 means the tag is stylistic and should be closed easily
620 * Depr: this element is deprecated
621 * DTD: 1 means that this element is valid only in the Loose DTD
622 * 2 means that this element is valid only in the Frameset DTD
624 * Name,Start Tag,End Tag,Save End,Empty,Deprecated,DTD,inline,Description
625 , subElements , impliedsubelt , Attributes, userdata
628 /* Definitions and a couple of vars for HTML Elements */
630 #define FONTSTYLE "tt", "i", "b", "u", "s", "strike", "big", "small"
631 #define NB_FONTSTYLE 8
632 #define PHRASE "em", "strong", "dfn", "code", "samp", "kbd", "var", "cite", "abbr", "acronym"
634 #define SPECIAL "a", "img", "applet", "embed", "object", "font", "basefont", "br", "script", "map", "q", "sub", "sup", "span", "bdo", "iframe"
635 #define NB_SPECIAL 16
636 #define INLINE FONTSTYLE, PHRASE, SPECIAL, FORMCTRL
637 #define NB_INLINE NB_PCDATA + NB_FONTSTYLE + NB_PHRASE + NB_SPECIAL + NB_FORMCTRL
638 #define BLOCK HEADING, LIST, "pre", "p", "dl", "div", "center", "noscript", "noframes", "blockquote", "form", "isindex", "hr", "table", "fieldset", "address"
639 #define NB_BLOCK NB_HEADING + NB_LIST + 14
640 #define FORMCTRL "input", "select", "textarea", "label", "button"
641 #define NB_FORMCTRL 5
644 #define HEADING "h1", "h2", "h3", "h4", "h5", "h6"
646 #define LIST "ul", "ol", "dir", "menu"
649 #define NB_MODIFIER 0
650 #define FLOW BLOCK,INLINE
651 #define NB_FLOW NB_BLOCK + NB_INLINE
655 static const char* const html_flow
[] = { FLOW
, NULL
} ;
656 static const char* const html_inline
[] = { INLINE
, NULL
} ;
658 /* placeholders: elts with content but no subelements */
659 static const char* const html_pcdata
[] = { NULL
} ;
660 #define html_cdata html_pcdata
663 /* ... and for HTML Attributes */
665 #define COREATTRS "id", "class", "style", "title"
666 #define NB_COREATTRS 4
667 #define I18N "lang", "dir"
669 #define EVENTS "onclick", "ondblclick", "onmousedown", "onmouseup", "onmouseover", "onmouseout", "onkeypress", "onkeydown", "onkeyup"
671 #define ATTRS COREATTRS,I18N,EVENTS
672 #define NB_ATTRS NB_NB_COREATTRS + NB_I18N + NB_EVENTS
673 #define CELLHALIGN "align", "char", "charoff"
674 #define NB_CELLHALIGN 3
675 #define CELLVALIGN "valign"
676 #define NB_CELLVALIGN 1
678 static const char* const html_attrs
[] = { ATTRS
, NULL
} ;
679 static const char* const core_i18n_attrs
[] = { COREATTRS
, I18N
, NULL
} ;
680 static const char* const core_attrs
[] = { COREATTRS
, NULL
} ;
681 static const char* const i18n_attrs
[] = { I18N
, NULL
} ;
684 /* Other declarations that should go inline ... */
685 static const char* const a_attrs
[] = { ATTRS
, "charset", "type", "name",
686 "href", "hreflang", "rel", "rev", "accesskey", "shape", "coords",
687 "tabindex", "onfocus", "onblur", NULL
} ;
688 static const char* const target_attr
[] = { "target", NULL
} ;
689 static const char* const rows_cols_attr
[] = { "rows", "cols", NULL
} ;
690 static const char* const alt_attr
[] = { "alt", NULL
} ;
691 static const char* const src_alt_attrs
[] = { "src", "alt", NULL
} ;
692 static const char* const href_attrs
[] = { "href", NULL
} ;
693 static const char* const clear_attrs
[] = { "clear", NULL
} ;
694 static const char* const inline_p
[] = { INLINE
, "p", NULL
} ;
696 static const char* const flow_param
[] = { FLOW
, "param", NULL
} ;
697 static const char* const applet_attrs
[] = { COREATTRS
, "codebase",
698 "archive", "alt", "name", "height", "width", "align",
699 "hspace", "vspace", NULL
} ;
700 static const char* const area_attrs
[] = { "shape", "coords", "href", "nohref",
701 "tabindex", "accesskey", "onfocus", "onblur", NULL
} ;
702 static const char* const basefont_attrs
[] =
703 { "id", "size", "color", "face", NULL
} ;
704 static const char* const quote_attrs
[] = { ATTRS
, "cite", NULL
} ;
705 static const char* const body_contents
[] = { FLOW
, "ins", "del", NULL
} ;
706 static const char* const body_attrs
[] = { ATTRS
, "onload", "onunload", NULL
} ;
707 static const char* const body_depr
[] = { "background", "bgcolor", "text",
708 "link", "vlink", "alink", NULL
} ;
709 static const char* const button_attrs
[] = { ATTRS
, "name", "value", "type",
710 "disabled", "tabindex", "accesskey", "onfocus", "onblur", NULL
} ;
713 static const char* const col_attrs
[] = { ATTRS
, "span", "width", CELLHALIGN
, CELLVALIGN
, NULL
} ;
714 static const char* const col_elt
[] = { "col", NULL
} ;
715 static const char* const edit_attrs
[] = { ATTRS
, "datetime", "cite", NULL
} ;
716 static const char* const compact_attrs
[] = { ATTRS
, "compact", NULL
} ;
717 static const char* const dl_contents
[] = { "dt", "dd", NULL
} ;
718 static const char* const compact_attr
[] = { "compact", NULL
} ;
719 static const char* const label_attr
[] = { "label", NULL
} ;
720 static const char* const fieldset_contents
[] = { FLOW
, "legend" } ;
721 static const char* const font_attrs
[] = { COREATTRS
, I18N
, "size", "color", "face" , NULL
} ;
722 static const char* const form_contents
[] = { HEADING
, LIST
, INLINE
, "pre", "p", "div", "center", "noscript", "noframes", "blockquote", "isindex", "hr", "table", "fieldset", "address", NULL
} ;
723 static const char* const form_attrs
[] = { ATTRS
, "method", "enctype", "accept", "name", "onsubmit", "onreset", "accept-charset", NULL
} ;
724 static const char* const frame_attrs
[] = { COREATTRS
, "longdesc", "name", "src", "frameborder", "marginwidth", "marginheight", "noresize", "scrolling" , NULL
} ;
725 static const char* const frameset_attrs
[] = { COREATTRS
, "rows", "cols", "onload", "onunload", NULL
} ;
726 static const char* const frameset_contents
[] = { "frameset", "frame", "noframes", NULL
} ;
727 static const char* const head_attrs
[] = { I18N
, "profile", NULL
} ;
728 static const char* const head_contents
[] = { "title", "isindex", "base", "script", "style", "meta", "link", "object", NULL
} ;
729 static const char* const hr_depr
[] = { "align", "noshade", "size", "width", NULL
} ;
730 static const char* const version_attr
[] = { "version", NULL
} ;
731 static const char* const html_content
[] = { "head", "body", "frameset", NULL
} ;
732 static const char* const iframe_attrs
[] = { COREATTRS
, "longdesc", "name", "src", "frameborder", "marginwidth", "marginheight", "scrolling", "align", "height", "width", NULL
} ;
733 static const char* const img_attrs
[] = { ATTRS
, "longdesc", "name", "height", "width", "usemap", "ismap", NULL
} ;
734 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
} ;
735 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
} ;
736 static const char* const prompt_attrs
[] = { COREATTRS
, I18N
, "prompt", NULL
} ;
737 static const char* const label_attrs
[] = { ATTRS
, "for", "accesskey", "onfocus", "onblur", NULL
} ;
738 static const char* const legend_attrs
[] = { ATTRS
, "accesskey", NULL
} ;
739 static const char* const align_attr
[] = { "align", NULL
} ;
740 static const char* const link_attrs
[] = { ATTRS
, "charset", "href", "hreflang", "type", "rel", "rev", "media", NULL
} ;
741 static const char* const map_contents
[] = { BLOCK
, "area", NULL
} ;
742 static const char* const name_attr
[] = { "name", NULL
} ;
743 static const char* const action_attr
[] = { "action", NULL
} ;
744 static const char* const blockli_elt
[] = { BLOCK
, "li", NULL
} ;
745 static const char* const meta_attrs
[] = { I18N
, "http-equiv", "name", "scheme", "charset", NULL
} ;
746 static const char* const content_attr
[] = { "content", NULL
} ;
747 static const char* const type_attr
[] = { "type", NULL
} ;
748 static const char* const noframes_content
[] = { "body", FLOW MODIFIER
, NULL
} ;
749 static const char* const object_contents
[] = { FLOW
, "param", NULL
} ;
750 static const char* const object_attrs
[] = { ATTRS
, "declare", "classid", "codebase", "data", "type", "codetype", "archive", "standby", "height", "width", "usemap", "name", "tabindex", NULL
} ;
751 static const char* const object_depr
[] = { "align", "border", "hspace", "vspace", NULL
} ;
752 static const char* const ol_attrs
[] = { "type", "compact", "start", NULL
} ;
753 static const char* const option_elt
[] = { "option", NULL
} ;
754 static const char* const optgroup_attrs
[] = { ATTRS
, "disabled", NULL
} ;
755 static const char* const option_attrs
[] = { ATTRS
, "disabled", "label", "selected", "value", NULL
} ;
756 static const char* const param_attrs
[] = { "id", "value", "valuetype", "type", NULL
} ;
757 static const char* const width_attr
[] = { "width", NULL
} ;
758 static const char* const pre_content
[] = { PHRASE
, "tt", "i", "b", "u", "s", "strike", "a", "br", "script", "map", "q", "span", "bdo", "iframe", NULL
} ;
759 static const char* const script_attrs
[] = { "charset", "src", "defer", "event", "for", NULL
} ;
760 static const char* const language_attr
[] = { "language", NULL
} ;
761 static const char* const select_content
[] = { "optgroup", "option", NULL
} ;
762 static const char* const select_attrs
[] = { ATTRS
, "name", "size", "multiple", "disabled", "tabindex", "onfocus", "onblur", "onchange", NULL
} ;
763 static const char* const style_attrs
[] = { I18N
, "media", "title", NULL
} ;
764 static const char* const table_attrs
[] = { ATTRS
, "summary", "width", "border", "frame", "rules", "cellspacing", "cellpadding", "datapagesize", NULL
} ;
765 static const char* const table_depr
[] = { "align", "bgcolor", NULL
} ;
766 static const char* const table_contents
[] = { "caption", "col", "colgroup", "thead", "tfoot", "tbody", "tr", NULL
} ;
767 static const char* const tr_elt
[] = { "tr", NULL
} ;
768 static const char* const talign_attrs
[] = { ATTRS
, CELLHALIGN
, CELLVALIGN
, NULL
} ;
769 static const char* const th_td_depr
[] = { "nowrap", "bgcolor", "width", "height", NULL
} ;
770 static const char* const th_td_attr
[] = { ATTRS
, "abbr", "axis", "headers", "scope", "rowspan", "colspan", CELLHALIGN
, CELLVALIGN
, NULL
} ;
771 static const char* const textarea_attrs
[] = { ATTRS
, "name", "disabled", "readonly", "tabindex", "accesskey", "onfocus", "onblur", "onselect", "onchange", NULL
} ;
772 static const char* const tr_contents
[] = { "th", "td", NULL
} ;
773 static const char* const bgcolor_attr
[] = { "bgcolor", NULL
} ;
774 static const char* const li_elt
[] = { "li", NULL
} ;
775 static const char* const ul_depr
[] = { "type", "compact", NULL
} ;
776 static const char* const dir_attr
[] = { "dir", NULL
} ;
778 #define DECL (const char**)
780 static const htmlElemDesc
781 html40ElementTable
[] = {
782 { "a", 0, 0, 0, 0, 0, 0, 1, "anchor ",
783 DECL html_inline
, NULL
, DECL a_attrs
, DECL target_attr
, NULL
785 { "abbr", 0, 0, 0, 0, 0, 0, 1, "abbreviated form",
786 DECL html_inline
, NULL
, DECL html_attrs
, NULL
, NULL
788 { "acronym", 0, 0, 0, 0, 0, 0, 1, "",
789 DECL html_inline
, NULL
, DECL html_attrs
, NULL
, NULL
791 { "address", 0, 0, 0, 0, 0, 0, 0, "information on author ",
792 DECL inline_p
, NULL
, DECL html_attrs
, NULL
, NULL
794 { "applet", 0, 0, 0, 0, 1, 1, 2, "java applet ",
795 DECL flow_param
, NULL
, NULL
, DECL applet_attrs
, NULL
797 { "area", 0, 2, 2, 1, 0, 0, 0, "client-side image map area ",
798 EMPTY
, NULL
, DECL area_attrs
, DECL target_attr
, DECL alt_attr
800 { "b", 0, 3, 0, 0, 0, 0, 1, "bold text style",
801 DECL html_inline
, NULL
, DECL html_attrs
, NULL
, NULL
803 { "base", 0, 2, 2, 1, 0, 0, 0, "document base uri ",
804 EMPTY
, NULL
, NULL
, DECL target_attr
, DECL href_attrs
806 { "basefont", 0, 2, 2, 1, 1, 1, 1, "base font size " ,
807 EMPTY
, NULL
, NULL
, DECL basefont_attrs
, NULL
809 { "bdo", 0, 0, 0, 0, 0, 0, 1, "i18n bidi over-ride ",
810 DECL html_inline
, NULL
, DECL core_i18n_attrs
, NULL
, DECL dir_attr
812 { "big", 0, 3, 0, 0, 0, 0, 1, "large text style",
813 DECL html_inline
, NULL
, DECL html_attrs
, NULL
, NULL
815 { "blockquote", 0, 0, 0, 0, 0, 0, 0, "long quotation ",
816 DECL html_flow
, NULL
, DECL quote_attrs
, NULL
, NULL
818 { "body", 1, 1, 0, 0, 0, 0, 0, "document body ",
819 DECL body_contents
, "div" , DECL body_attrs
, DECL body_depr
, NULL
821 { "br", 0, 2, 2, 1, 0, 0, 1, "forced line break ",
822 EMPTY
, NULL
, DECL core_attrs
, DECL clear_attrs
, NULL
824 { "button", 0, 0, 0, 0, 0, 0, 2, "push button ",
825 DECL html_flow MODIFIER
, NULL
, DECL button_attrs
, NULL
, NULL
827 { "caption", 0, 0, 0, 0, 0, 0, 0, "table caption ",
828 DECL html_inline
, NULL
, DECL html_attrs
, NULL
, NULL
830 { "center", 0, 3, 0, 0, 1, 1, 0, "shorthand for div align=center ",
831 DECL html_flow
, NULL
, NULL
, DECL html_attrs
, NULL
833 { "cite", 0, 0, 0, 0, 0, 0, 1, "citation",
834 DECL html_inline
, NULL
, DECL html_attrs
, NULL
, NULL
836 { "code", 0, 0, 0, 0, 0, 0, 1, "computer code fragment",
837 DECL html_inline
, NULL
, DECL html_attrs
, NULL
, NULL
839 { "col", 0, 2, 2, 1, 0, 0, 0, "table column ",
840 EMPTY
, NULL
, DECL col_attrs
, NULL
, NULL
842 { "colgroup", 0, 1, 0, 0, 0, 0, 0, "table column group ",
843 DECL col_elt
, "col" , DECL col_attrs
, NULL
, NULL
845 { "dd", 0, 1, 0, 0, 0, 0, 0, "definition description ",
846 DECL html_flow
, NULL
, DECL html_attrs
, NULL
, NULL
848 { "del", 0, 0, 0, 0, 0, 0, 2, "deleted text ",
849 DECL html_flow
, NULL
, DECL edit_attrs
, NULL
, NULL
851 { "dfn", 0, 0, 0, 0, 0, 0, 1, "instance definition",
852 DECL html_inline
, NULL
, DECL html_attrs
, NULL
, NULL
854 { "dir", 0, 0, 0, 0, 1, 1, 0, "directory list",
855 DECL blockli_elt
, "li" , NULL
, DECL compact_attrs
, NULL
857 { "div", 0, 0, 0, 0, 0, 0, 0, "generic language/style container",
858 DECL html_flow
, NULL
, DECL html_attrs
, DECL align_attr
, NULL
860 { "dl", 0, 0, 0, 0, 0, 0, 0, "definition list ",
861 DECL dl_contents
, "dd" , DECL html_attrs
, DECL compact_attr
, NULL
863 { "dt", 0, 1, 0, 0, 0, 0, 0, "definition term ",
864 DECL html_inline
, NULL
, DECL html_attrs
, NULL
, NULL
866 { "em", 0, 3, 0, 0, 0, 0, 1, "emphasis",
867 DECL html_inline
, NULL
, DECL html_attrs
, NULL
, NULL
869 { "embed", 0, 1, 0, 0, 1, 1, 1, "generic embedded object ",
870 EMPTY
, NULL
, DECL embed_attrs
, NULL
, NULL
872 { "fieldset", 0, 0, 0, 0, 0, 0, 0, "form control group ",
873 DECL fieldset_contents
, NULL
, DECL html_attrs
, NULL
, NULL
875 { "font", 0, 3, 0, 0, 1, 1, 1, "local change to font ",
876 DECL html_inline
, NULL
, NULL
, DECL font_attrs
, NULL
878 { "form", 0, 0, 0, 0, 0, 0, 0, "interactive form ",
879 DECL form_contents
, "fieldset", DECL form_attrs
, DECL target_attr
, DECL action_attr
881 { "frame", 0, 2, 2, 1, 0, 2, 0, "subwindow " ,
882 EMPTY
, NULL
, NULL
, DECL frame_attrs
, NULL
884 { "frameset", 0, 0, 0, 0, 0, 2, 0, "window subdivision" ,
885 DECL frameset_contents
, "noframes" , NULL
, DECL frameset_attrs
, NULL
887 { "h1", 0, 0, 0, 0, 0, 0, 0, "heading ",
888 DECL html_inline
, NULL
, DECL html_attrs
, DECL align_attr
, NULL
890 { "h2", 0, 0, 0, 0, 0, 0, 0, "heading ",
891 DECL html_inline
, NULL
, DECL html_attrs
, DECL align_attr
, NULL
893 { "h3", 0, 0, 0, 0, 0, 0, 0, "heading ",
894 DECL html_inline
, NULL
, DECL html_attrs
, DECL align_attr
, NULL
896 { "h4", 0, 0, 0, 0, 0, 0, 0, "heading ",
897 DECL html_inline
, NULL
, DECL html_attrs
, DECL align_attr
, NULL
899 { "h5", 0, 0, 0, 0, 0, 0, 0, "heading ",
900 DECL html_inline
, NULL
, DECL html_attrs
, DECL align_attr
, NULL
902 { "h6", 0, 0, 0, 0, 0, 0, 0, "heading ",
903 DECL html_inline
, NULL
, DECL html_attrs
, DECL align_attr
, NULL
905 { "head", 1, 1, 0, 0, 0, 0, 0, "document head ",
906 DECL head_contents
, NULL
, DECL head_attrs
, NULL
, NULL
908 { "hr", 0, 2, 2, 1, 0, 0, 0, "horizontal rule " ,
909 EMPTY
, NULL
, DECL html_attrs
, DECL hr_depr
, NULL
911 { "html", 1, 1, 0, 0, 0, 0, 0, "document root element ",
912 DECL html_content
, NULL
, DECL i18n_attrs
, DECL version_attr
, NULL
914 { "i", 0, 3, 0, 0, 0, 0, 1, "italic text style",
915 DECL html_inline
, NULL
, DECL html_attrs
, NULL
, NULL
917 { "iframe", 0, 0, 0, 0, 0, 1, 2, "inline subwindow ",
918 DECL html_flow
, NULL
, NULL
, DECL iframe_attrs
, NULL
920 { "img", 0, 2, 2, 1, 0, 0, 1, "embedded image ",
921 EMPTY
, NULL
, DECL img_attrs
, DECL align_attr
, DECL src_alt_attrs
923 { "input", 0, 2, 2, 1, 0, 0, 1, "form control ",
924 EMPTY
, NULL
, DECL input_attrs
, DECL align_attr
, NULL
926 { "ins", 0, 0, 0, 0, 0, 0, 2, "inserted text",
927 DECL html_flow
, NULL
, DECL edit_attrs
, NULL
, NULL
929 { "isindex", 0, 2, 2, 1, 1, 1, 0, "single line prompt ",
930 EMPTY
, NULL
, NULL
, DECL prompt_attrs
, NULL
932 { "kbd", 0, 0, 0, 0, 0, 0, 1, "text to be entered by the user",
933 DECL html_inline
, NULL
, DECL html_attrs
, NULL
, NULL
935 { "label", 0, 0, 0, 0, 0, 0, 1, "form field label text ",
936 DECL html_inline MODIFIER
, NULL
, DECL label_attrs
, NULL
, NULL
938 { "legend", 0, 0, 0, 0, 0, 0, 0, "fieldset legend ",
939 DECL html_inline
, NULL
, DECL legend_attrs
, DECL align_attr
, NULL
941 { "li", 0, 1, 1, 0, 0, 0, 0, "list item ",
942 DECL html_flow
, NULL
, DECL html_attrs
, NULL
, NULL
944 { "link", 0, 2, 2, 1, 0, 0, 0, "a media-independent link ",
945 EMPTY
, NULL
, DECL link_attrs
, DECL target_attr
, NULL
947 { "map", 0, 0, 0, 0, 0, 0, 2, "client-side image map ",
948 DECL map_contents
, NULL
, DECL html_attrs
, NULL
, DECL name_attr
950 { "menu", 0, 0, 0, 0, 1, 1, 0, "menu list ",
951 DECL blockli_elt
, NULL
, NULL
, DECL compact_attrs
, NULL
953 { "meta", 0, 2, 2, 1, 0, 0, 0, "generic metainformation ",
954 EMPTY
, NULL
, DECL meta_attrs
, NULL
, DECL content_attr
956 { "noframes", 0, 0, 0, 0, 0, 2, 0, "alternate content container for non frame-based rendering ",
957 DECL noframes_content
, "body" , DECL html_attrs
, NULL
, NULL
959 { "noscript", 0, 0, 0, 0, 0, 0, 0, "alternate content container for non script-based rendering ",
960 DECL html_flow
, "div", DECL html_attrs
, NULL
, NULL
962 { "object", 0, 0, 0, 0, 0, 0, 2, "generic embedded object ",
963 DECL object_contents
, "div" , DECL object_attrs
, DECL object_depr
, NULL
965 { "ol", 0, 0, 0, 0, 0, 0, 0, "ordered list ",
966 DECL li_elt
, "li" , DECL html_attrs
, DECL ol_attrs
, NULL
968 { "optgroup", 0, 0, 0, 0, 0, 0, 0, "option group ",
969 DECL option_elt
, "option", DECL optgroup_attrs
, NULL
, DECL label_attr
971 { "option", 0, 1, 0, 0, 0, 0, 0, "selectable choice " ,
972 DECL html_pcdata
, NULL
, DECL option_attrs
, NULL
, NULL
974 { "p", 0, 1, 0, 0, 0, 0, 0, "paragraph ",
975 DECL html_inline
, NULL
, DECL html_attrs
, DECL align_attr
, NULL
977 { "param", 0, 2, 2, 1, 0, 0, 0, "named property value ",
978 EMPTY
, NULL
, DECL param_attrs
, NULL
, DECL name_attr
980 { "pre", 0, 0, 0, 0, 0, 0, 0, "preformatted text ",
981 DECL pre_content
, NULL
, DECL html_attrs
, DECL width_attr
, NULL
983 { "q", 0, 0, 0, 0, 0, 0, 1, "short inline quotation ",
984 DECL html_inline
, NULL
, DECL quote_attrs
, NULL
, NULL
986 { "s", 0, 3, 0, 0, 1, 1, 1, "strike-through text style",
987 DECL html_inline
, NULL
, NULL
, DECL html_attrs
, NULL
989 { "samp", 0, 0, 0, 0, 0, 0, 1, "sample program output, scripts, etc.",
990 DECL html_inline
, NULL
, DECL html_attrs
, NULL
, NULL
992 { "script", 0, 0, 0, 0, 0, 0, 2, "script statements ",
993 DECL html_cdata
, NULL
, DECL script_attrs
, DECL language_attr
, DECL type_attr
995 { "select", 0, 0, 0, 0, 0, 0, 1, "option selector ",
996 DECL select_content
, NULL
, DECL select_attrs
, NULL
, NULL
998 { "small", 0, 3, 0, 0, 0, 0, 1, "small text style",
999 DECL html_inline
, NULL
, DECL html_attrs
, NULL
, NULL
1001 { "span", 0, 0, 0, 0, 0, 0, 1, "generic language/style container ",
1002 DECL html_inline
, NULL
, DECL html_attrs
, NULL
, NULL
1004 { "strike", 0, 3, 0, 0, 1, 1, 1, "strike-through text",
1005 DECL html_inline
, NULL
, NULL
, DECL html_attrs
, NULL
1007 { "strong", 0, 3, 0, 0, 0, 0, 1, "strong emphasis",
1008 DECL html_inline
, NULL
, DECL html_attrs
, NULL
, NULL
1010 { "style", 0, 0, 0, 0, 0, 0, 0, "style info ",
1011 DECL html_cdata
, NULL
, DECL style_attrs
, NULL
, DECL type_attr
1013 { "sub", 0, 3, 0, 0, 0, 0, 1, "subscript",
1014 DECL html_inline
, NULL
, DECL html_attrs
, NULL
, NULL
1016 { "sup", 0, 3, 0, 0, 0, 0, 1, "superscript ",
1017 DECL html_inline
, NULL
, DECL html_attrs
, NULL
, NULL
1019 { "table", 0, 0, 0, 0, 0, 0, 0, "",
1020 DECL table_contents
, "tr" , DECL table_attrs
, DECL table_depr
, NULL
1022 { "tbody", 1, 0, 0, 0, 0, 0, 0, "table body ",
1023 DECL tr_elt
, "tr" , DECL talign_attrs
, NULL
, NULL
1025 { "td", 0, 0, 0, 0, 0, 0, 0, "table data cell",
1026 DECL html_flow
, NULL
, DECL th_td_attr
, DECL th_td_depr
, NULL
1028 { "textarea", 0, 0, 0, 0, 0, 0, 1, "multi-line text field ",
1029 DECL html_pcdata
, NULL
, DECL textarea_attrs
, NULL
, DECL rows_cols_attr
1031 { "tfoot", 0, 1, 0, 0, 0, 0, 0, "table footer ",
1032 DECL tr_elt
, "tr" , DECL talign_attrs
, NULL
, NULL
1034 { "th", 0, 1, 0, 0, 0, 0, 0, "table header cell",
1035 DECL html_flow
, NULL
, DECL th_td_attr
, DECL th_td_depr
, NULL
1037 { "thead", 0, 1, 0, 0, 0, 0, 0, "table header ",
1038 DECL tr_elt
, "tr" , DECL talign_attrs
, NULL
, NULL
1040 { "title", 0, 0, 0, 0, 0, 0, 0, "document title ",
1041 DECL html_pcdata
, NULL
, DECL i18n_attrs
, NULL
, NULL
1043 { "tr", 0, 0, 0, 0, 0, 0, 0, "table row ",
1044 DECL tr_contents
, "td" , DECL talign_attrs
, DECL bgcolor_attr
, NULL
1046 { "tt", 0, 3, 0, 0, 0, 0, 1, "teletype or monospaced text style",
1047 DECL html_inline
, NULL
, DECL html_attrs
, NULL
, NULL
1049 { "u", 0, 3, 0, 0, 1, 1, 1, "underlined text style",
1050 DECL html_inline
, NULL
, NULL
, DECL html_attrs
, NULL
1052 { "ul", 0, 0, 0, 0, 0, 0, 0, "unordered list ",
1053 DECL li_elt
, "li" , DECL html_attrs
, DECL ul_depr
, NULL
1055 { "var", 0, 0, 0, 0, 0, 0, 1, "instance of a variable or program argument",
1056 DECL html_inline
, NULL
, DECL html_attrs
, NULL
, NULL
1063 } htmlStartCloseEntry
;
1066 * start tags that imply the end of current element
1068 static const htmlStartCloseEntry htmlStartClose
[] = {
1070 { "a", "fieldset" },
1074 { "address", "dd" },
1075 { "address", "dl" },
1076 { "address", "dt" },
1077 { "address", "form" },
1078 { "address", "li" },
1079 { "address", "ul" },
1085 { "caption", "col" },
1086 { "caption", "colgroup" },
1087 { "caption", "tbody" },
1088 { "caption", "tfoot" },
1089 { "caption", "thead" },
1090 { "caption", "tr" },
1092 { "col", "colgroup" },
1097 { "colgroup", "colgroup" },
1098 { "colgroup", "tbody" },
1099 { "colgroup", "tfoot" },
1100 { "colgroup", "thead" },
1101 { "colgroup", "tr" },
1112 { "font", "center" },
1116 { "h1", "fieldset" },
1121 { "h2", "fieldset" },
1126 { "h3", "fieldset" },
1131 { "h4", "fieldset" },
1136 { "h5", "fieldset" },
1141 { "h6", "fieldset" },
1148 { "head", "acronym" },
1149 { "head", "address" },
1153 { "head", "blockquote" },
1156 { "head", "center" },
1166 { "head", "fieldset" },
1169 { "head", "frameset" },
1178 { "head", "iframe" },
1182 { "head", "listing" },
1191 { "head", "small" },
1193 { "head", "strike" },
1194 { "head", "strong" },
1197 { "head", "table" },
1208 { "legend", "fieldset" },
1211 { "link", "frameset" },
1212 { "listing", "dd" },
1213 { "listing", "dl" },
1214 { "listing", "dt" },
1215 { "listing", "fieldset" },
1216 { "listing", "form" },
1217 { "listing", "li" },
1218 { "listing", "table" },
1219 { "listing", "ul" },
1227 { "option", "optgroup" },
1228 { "option", "option" },
1230 { "p", "blockquote" },
1235 { "p", "colgroup" },
1241 { "p", "fieldset" },
1243 { "p", "frameset" },
1270 { "pre", "fieldset" },
1276 { "script", "noscript" },
1281 { "style", "body" },
1282 { "style", "frameset" },
1283 { "tbody", "tbody" },
1284 { "tbody", "tfoot" },
1290 { "tfoot", "tbody" },
1296 { "thead", "tbody" },
1297 { "thead", "tfoot" },
1298 { "title", "body" },
1299 { "title", "frameset" },
1307 { "ul", "address" },
1315 { "xmp", "fieldset" },
1323 * The list of HTML elements which are supposed not to have
1324 * CDATA content and where a p element will be implied
1326 * TODO: extend that list by reading the HTML SGML DTD on
1329 static const char *const htmlNoContentElements
[] = {
1336 * The list of HTML attributes which are of content %Script;
1337 * NOTE: when adding ones, check htmlIsScriptAttribute() since
1338 * it assumes the name starts with 'on'
1340 static const char *const htmlScriptAttributes
[] = {
1362 * This table is used by the htmlparser to know what to do with
1363 * broken html pages. By assigning different priorities to different
1364 * elements the parser can decide how to handle extra endtags.
1365 * Endtags are only allowed to close elements with lower or equal
1374 static const elementPriority htmlEndPriority
[] = {
1386 {NULL
, 100} /* Default priority */
1389 /************************************************************************
1391 * functions to handle HTML specific data *
1393 ************************************************************************/
1396 * htmlInitAutoClose:
1398 * DEPRECATED: This function will be made private. Call xmlInitParser to
1399 * initialize the library.
1401 * This is a no-op now.
1404 htmlInitAutoClose(void) {
1408 htmlCompareTags(const void *key
, const void *member
) {
1409 const xmlChar
*tag
= (const xmlChar
*) key
;
1410 const htmlElemDesc
*desc
= (const htmlElemDesc
*) member
;
1412 return(xmlStrcasecmp(tag
, BAD_CAST desc
->name
));
1417 * @tag: The tag name in lowercase
1419 * Lookup the HTML tag in the ElementTable
1421 * Returns the related htmlElemDescPtr or NULL if not found.
1423 const htmlElemDesc
*
1424 htmlTagLookup(const xmlChar
*tag
) {
1428 return((const htmlElemDesc
*) bsearch(tag
, html40ElementTable
,
1429 sizeof(html40ElementTable
) / sizeof(htmlElemDesc
),
1430 sizeof(htmlElemDesc
), htmlCompareTags
));
1434 * htmlGetEndPriority:
1435 * @name: The name of the element to look up the priority for.
1437 * Return value: The "endtag" priority.
1440 htmlGetEndPriority (const xmlChar
*name
) {
1443 while ((htmlEndPriority
[i
].name
!= NULL
) &&
1444 (!xmlStrEqual((const xmlChar
*)htmlEndPriority
[i
].name
, name
)))
1447 return(htmlEndPriority
[i
].priority
);
1452 htmlCompareStartClose(const void *vkey
, const void *member
) {
1453 const htmlStartCloseEntry
*key
= (const htmlStartCloseEntry
*) vkey
;
1454 const htmlStartCloseEntry
*entry
= (const htmlStartCloseEntry
*) member
;
1457 ret
= strcmp(key
->oldTag
, entry
->oldTag
);
1459 ret
= strcmp(key
->newTag
, entry
->newTag
);
1465 * htmlCheckAutoClose:
1466 * @newtag: The new tag name
1467 * @oldtag: The old tag name
1469 * Checks whether the new tag is one of the registered valid tags for
1472 * Returns 0 if no, 1 if yes.
1475 htmlCheckAutoClose(const xmlChar
* newtag
, const xmlChar
* oldtag
)
1477 htmlStartCloseEntry key
;
1480 key
.oldTag
= (const char *) oldtag
;
1481 key
.newTag
= (const char *) newtag
;
1482 res
= bsearch(&key
, htmlStartClose
,
1483 sizeof(htmlStartClose
) / sizeof(htmlStartCloseEntry
),
1484 sizeof(htmlStartCloseEntry
), htmlCompareStartClose
);
1485 return(res
!= NULL
);
1489 * htmlAutoCloseOnClose:
1490 * @ctxt: an HTML parser context
1491 * @newtag: The new tag name
1492 * @force: force the tag closure
1494 * The HTML DTD allows an ending tag to implicitly close other tags.
1497 htmlAutoCloseOnClose(htmlParserCtxtPtr ctxt
, const xmlChar
* newtag
)
1499 const htmlElemDesc
*info
;
1502 priority
= htmlGetEndPriority(newtag
);
1504 for (i
= (ctxt
->nameNr
- 1); i
>= 0; i
--) {
1506 if (xmlStrEqual(newtag
, ctxt
->nameTab
[i
]))
1509 * A misplaced endtag can only close elements with lower
1510 * or equal priority, so if we find an element with higher
1511 * priority before we find an element with
1512 * matching name, we just ignore this endtag
1514 if (htmlGetEndPriority(ctxt
->nameTab
[i
]) > priority
)
1520 while (!xmlStrEqual(newtag
, ctxt
->name
)) {
1521 info
= htmlTagLookup(ctxt
->name
);
1522 if ((info
!= NULL
) && (info
->endTag
== 3)) {
1523 htmlParseErr(ctxt
, XML_ERR_TAG_NAME_MISMATCH
,
1524 "Opening and ending tag mismatch: %s and %s\n",
1525 newtag
, ctxt
->name
);
1527 if ((ctxt
->sax
!= NULL
) && (ctxt
->sax
->endElement
!= NULL
))
1528 ctxt
->sax
->endElement(ctxt
->userData
, ctxt
->name
);
1534 * htmlAutoCloseOnEnd:
1535 * @ctxt: an HTML parser context
1537 * Close all remaining tags at the end of the stream
1540 htmlAutoCloseOnEnd(htmlParserCtxtPtr ctxt
)
1544 if (ctxt
->nameNr
== 0)
1546 for (i
= (ctxt
->nameNr
- 1); i
>= 0; i
--) {
1547 if ((ctxt
->sax
!= NULL
) && (ctxt
->sax
->endElement
!= NULL
))
1548 ctxt
->sax
->endElement(ctxt
->userData
, ctxt
->name
);
1555 * @ctxt: an HTML parser context
1556 * @newtag: The new tag name or NULL
1558 * The HTML DTD allows a tag to implicitly close other tags.
1559 * The list is kept in htmlStartClose array. This function is
1560 * called when a new tag has been detected and generates the
1561 * appropriates closes if possible/needed.
1562 * If newtag is NULL this mean we are at the end of the resource
1563 * and we should check
1566 htmlAutoClose(htmlParserCtxtPtr ctxt
, const xmlChar
* newtag
)
1568 while ((newtag
!= NULL
) && (ctxt
->name
!= NULL
) &&
1569 (htmlCheckAutoClose(newtag
, ctxt
->name
))) {
1570 if ((ctxt
->sax
!= NULL
) && (ctxt
->sax
->endElement
!= NULL
))
1571 ctxt
->sax
->endElement(ctxt
->userData
, ctxt
->name
);
1574 if (newtag
== NULL
) {
1575 htmlAutoCloseOnEnd(ctxt
);
1578 while ((newtag
== NULL
) && (ctxt
->name
!= NULL
) &&
1579 ((xmlStrEqual(ctxt
->name
, BAD_CAST
"head")) ||
1580 (xmlStrEqual(ctxt
->name
, BAD_CAST
"body")) ||
1581 (xmlStrEqual(ctxt
->name
, BAD_CAST
"html")))) {
1582 if ((ctxt
->sax
!= NULL
) && (ctxt
->sax
->endElement
!= NULL
))
1583 ctxt
->sax
->endElement(ctxt
->userData
, ctxt
->name
);
1590 * @doc: the HTML document
1591 * @name: The tag name
1592 * @elem: the HTML element
1594 * The HTML DTD allows a tag to implicitly close other tags.
1595 * The list is kept in htmlStartClose array. This function checks
1596 * if the element or one of it's children would autoclose the
1599 * Returns 1 if autoclose, 0 otherwise
1602 htmlAutoCloseTag(htmlDocPtr doc
, const xmlChar
*name
, htmlNodePtr elem
) {
1605 if (elem
== NULL
) return(1);
1606 if (xmlStrEqual(name
, elem
->name
)) return(0);
1607 if (htmlCheckAutoClose(elem
->name
, name
)) return(1);
1608 child
= elem
->children
;
1609 while (child
!= NULL
) {
1610 if (htmlAutoCloseTag(doc
, name
, child
)) return(1);
1611 child
= child
->next
;
1618 * @doc: the HTML document
1619 * @elem: the HTML element
1621 * The HTML DTD allows a tag to implicitly close other tags.
1622 * The list is kept in htmlStartClose array. This function checks
1623 * if a tag is autoclosed by one of it's child
1625 * Returns 1 if autoclosed, 0 otherwise
1628 htmlIsAutoClosed(htmlDocPtr doc
, htmlNodePtr elem
) {
1631 if (elem
== NULL
) return(1);
1632 child
= elem
->children
;
1633 while (child
!= NULL
) {
1634 if (htmlAutoCloseTag(doc
, elem
->name
, child
)) return(1);
1635 child
= child
->next
;
1642 * @ctxt: an HTML parser context
1643 * @newtag: The new tag name
1645 * The HTML DTD allows a tag to exists only implicitly
1646 * called when a new tag has been detected and generates the
1647 * appropriates implicit tags if missing
1650 htmlCheckImplied(htmlParserCtxtPtr ctxt
, const xmlChar
*newtag
) {
1653 if (ctxt
->options
& HTML_PARSE_NOIMPLIED
)
1655 if (!htmlOmittedDefaultValue
)
1657 if (xmlStrEqual(newtag
, BAD_CAST
"html"))
1659 if (ctxt
->nameNr
<= 0) {
1660 htmlnamePush(ctxt
, BAD_CAST
"html");
1661 if ((ctxt
->sax
!= NULL
) && (ctxt
->sax
->startElement
!= NULL
))
1662 ctxt
->sax
->startElement(ctxt
->userData
, BAD_CAST
"html", NULL
);
1664 if ((xmlStrEqual(newtag
, BAD_CAST
"body")) || (xmlStrEqual(newtag
, BAD_CAST
"head")))
1666 if ((ctxt
->nameNr
<= 1) &&
1667 ((xmlStrEqual(newtag
, BAD_CAST
"script")) ||
1668 (xmlStrEqual(newtag
, BAD_CAST
"style")) ||
1669 (xmlStrEqual(newtag
, BAD_CAST
"meta")) ||
1670 (xmlStrEqual(newtag
, BAD_CAST
"link")) ||
1671 (xmlStrEqual(newtag
, BAD_CAST
"title")) ||
1672 (xmlStrEqual(newtag
, BAD_CAST
"base")))) {
1673 if (ctxt
->html
>= 3) {
1674 /* we already saw or generated an <head> before */
1678 * dropped OBJECT ... i you put it first BODY will be
1681 htmlnamePush(ctxt
, BAD_CAST
"head");
1682 if ((ctxt
->sax
!= NULL
) && (ctxt
->sax
->startElement
!= NULL
))
1683 ctxt
->sax
->startElement(ctxt
->userData
, BAD_CAST
"head", NULL
);
1684 } else if ((!xmlStrEqual(newtag
, BAD_CAST
"noframes")) &&
1685 (!xmlStrEqual(newtag
, BAD_CAST
"frame")) &&
1686 (!xmlStrEqual(newtag
, BAD_CAST
"frameset"))) {
1687 if (ctxt
->html
>= 10) {
1688 /* we already saw or generated a <body> before */
1691 for (i
= 0;i
< ctxt
->nameNr
;i
++) {
1692 if (xmlStrEqual(ctxt
->nameTab
[i
], BAD_CAST
"body")) {
1695 if (xmlStrEqual(ctxt
->nameTab
[i
], BAD_CAST
"head")) {
1700 htmlnamePush(ctxt
, BAD_CAST
"body");
1701 if ((ctxt
->sax
!= NULL
) && (ctxt
->sax
->startElement
!= NULL
))
1702 ctxt
->sax
->startElement(ctxt
->userData
, BAD_CAST
"body", NULL
);
1707 * htmlCheckParagraph
1708 * @ctxt: an HTML parser context
1710 * Check whether a p element need to be implied before inserting
1711 * characters in the current element.
1713 * Returns 1 if a paragraph has been inserted, 0 if not and -1
1718 htmlCheckParagraph(htmlParserCtxtPtr ctxt
) {
1726 htmlAutoClose(ctxt
, BAD_CAST
"p");
1727 htmlCheckImplied(ctxt
, BAD_CAST
"p");
1728 htmlnamePush(ctxt
, BAD_CAST
"p");
1729 if ((ctxt
->sax
!= NULL
) && (ctxt
->sax
->startElement
!= NULL
))
1730 ctxt
->sax
->startElement(ctxt
->userData
, BAD_CAST
"p", NULL
);
1733 if (!htmlOmittedDefaultValue
)
1735 for (i
= 0; htmlNoContentElements
[i
] != NULL
; i
++) {
1736 if (xmlStrEqual(tag
, BAD_CAST htmlNoContentElements
[i
])) {
1737 htmlAutoClose(ctxt
, BAD_CAST
"p");
1738 htmlCheckImplied(ctxt
, BAD_CAST
"p");
1739 htmlnamePush(ctxt
, BAD_CAST
"p");
1740 if ((ctxt
->sax
!= NULL
) && (ctxt
->sax
->startElement
!= NULL
))
1741 ctxt
->sax
->startElement(ctxt
->userData
, BAD_CAST
"p", NULL
);
1749 * htmlIsScriptAttribute:
1750 * @name: an attribute name
1752 * Check if an attribute is of content type Script
1754 * Returns 1 is the attribute is a script 0 otherwise
1757 htmlIsScriptAttribute(const xmlChar
*name
) {
1763 * all script attributes start with 'on'
1765 if ((name
[0] != 'o') || (name
[1] != 'n'))
1768 i
< sizeof(htmlScriptAttributes
)/sizeof(htmlScriptAttributes
[0]);
1770 if (xmlStrEqual(name
, (const xmlChar
*) htmlScriptAttributes
[i
]))
1776 /************************************************************************
1778 * The list of HTML predefined entities *
1780 ************************************************************************/
1783 static const htmlEntityDesc html40EntitiesTable
[] = {
1785 * the 4 absolute ones, plus apostrophe.
1787 { 34, "quot", "quotation mark = APL quote, U+0022 ISOnum" },
1788 { 38, "amp", "ampersand, U+0026 ISOnum" },
1789 { 39, "apos", "single quote" },
1790 { 60, "lt", "less-than sign, U+003C ISOnum" },
1791 { 62, "gt", "greater-than sign, U+003E ISOnum" },
1794 * A bunch still in the 128-255 range
1795 * Replacing them depend really on the charset used.
1797 { 160, "nbsp", "no-break space = non-breaking space, U+00A0 ISOnum" },
1798 { 161, "iexcl","inverted exclamation mark, U+00A1 ISOnum" },
1799 { 162, "cent", "cent sign, U+00A2 ISOnum" },
1800 { 163, "pound","pound sign, U+00A3 ISOnum" },
1801 { 164, "curren","currency sign, U+00A4 ISOnum" },
1802 { 165, "yen", "yen sign = yuan sign, U+00A5 ISOnum" },
1803 { 166, "brvbar","broken bar = broken vertical bar, U+00A6 ISOnum" },
1804 { 167, "sect", "section sign, U+00A7 ISOnum" },
1805 { 168, "uml", "diaeresis = spacing diaeresis, U+00A8 ISOdia" },
1806 { 169, "copy", "copyright sign, U+00A9 ISOnum" },
1807 { 170, "ordf", "feminine ordinal indicator, U+00AA ISOnum" },
1808 { 171, "laquo","left-pointing double angle quotation mark = left pointing guillemet, U+00AB ISOnum" },
1809 { 172, "not", "not sign, U+00AC ISOnum" },
1810 { 173, "shy", "soft hyphen = discretionary hyphen, U+00AD ISOnum" },
1811 { 174, "reg", "registered sign = registered trade mark sign, U+00AE ISOnum" },
1812 { 175, "macr", "macron = spacing macron = overline = APL overbar, U+00AF ISOdia" },
1813 { 176, "deg", "degree sign, U+00B0 ISOnum" },
1814 { 177, "plusmn","plus-minus sign = plus-or-minus sign, U+00B1 ISOnum" },
1815 { 178, "sup2", "superscript two = superscript digit two = squared, U+00B2 ISOnum" },
1816 { 179, "sup3", "superscript three = superscript digit three = cubed, U+00B3 ISOnum" },
1817 { 180, "acute","acute accent = spacing acute, U+00B4 ISOdia" },
1818 { 181, "micro","micro sign, U+00B5 ISOnum" },
1819 { 182, "para", "pilcrow sign = paragraph sign, U+00B6 ISOnum" },
1820 { 183, "middot","middle dot = Georgian comma Greek middle dot, U+00B7 ISOnum" },
1821 { 184, "cedil","cedilla = spacing cedilla, U+00B8 ISOdia" },
1822 { 185, "sup1", "superscript one = superscript digit one, U+00B9 ISOnum" },
1823 { 186, "ordm", "masculine ordinal indicator, U+00BA ISOnum" },
1824 { 187, "raquo","right-pointing double angle quotation mark right pointing guillemet, U+00BB ISOnum" },
1825 { 188, "frac14","vulgar fraction one quarter = fraction one quarter, U+00BC ISOnum" },
1826 { 189, "frac12","vulgar fraction one half = fraction one half, U+00BD ISOnum" },
1827 { 190, "frac34","vulgar fraction three quarters = fraction three quarters, U+00BE ISOnum" },
1828 { 191, "iquest","inverted question mark = turned question mark, U+00BF ISOnum" },
1829 { 192, "Agrave","latin capital letter A with grave = latin capital letter A grave, U+00C0 ISOlat1" },
1830 { 193, "Aacute","latin capital letter A with acute, U+00C1 ISOlat1" },
1831 { 194, "Acirc","latin capital letter A with circumflex, U+00C2 ISOlat1" },
1832 { 195, "Atilde","latin capital letter A with tilde, U+00C3 ISOlat1" },
1833 { 196, "Auml", "latin capital letter A with diaeresis, U+00C4 ISOlat1" },
1834 { 197, "Aring","latin capital letter A with ring above = latin capital letter A ring, U+00C5 ISOlat1" },
1835 { 198, "AElig","latin capital letter AE = latin capital ligature AE, U+00C6 ISOlat1" },
1836 { 199, "Ccedil","latin capital letter C with cedilla, U+00C7 ISOlat1" },
1837 { 200, "Egrave","latin capital letter E with grave, U+00C8 ISOlat1" },
1838 { 201, "Eacute","latin capital letter E with acute, U+00C9 ISOlat1" },
1839 { 202, "Ecirc","latin capital letter E with circumflex, U+00CA ISOlat1" },
1840 { 203, "Euml", "latin capital letter E with diaeresis, U+00CB ISOlat1" },
1841 { 204, "Igrave","latin capital letter I with grave, U+00CC ISOlat1" },
1842 { 205, "Iacute","latin capital letter I with acute, U+00CD ISOlat1" },
1843 { 206, "Icirc","latin capital letter I with circumflex, U+00CE ISOlat1" },
1844 { 207, "Iuml", "latin capital letter I with diaeresis, U+00CF ISOlat1" },
1845 { 208, "ETH", "latin capital letter ETH, U+00D0 ISOlat1" },
1846 { 209, "Ntilde","latin capital letter N with tilde, U+00D1 ISOlat1" },
1847 { 210, "Ograve","latin capital letter O with grave, U+00D2 ISOlat1" },
1848 { 211, "Oacute","latin capital letter O with acute, U+00D3 ISOlat1" },
1849 { 212, "Ocirc","latin capital letter O with circumflex, U+00D4 ISOlat1" },
1850 { 213, "Otilde","latin capital letter O with tilde, U+00D5 ISOlat1" },
1851 { 214, "Ouml", "latin capital letter O with diaeresis, U+00D6 ISOlat1" },
1852 { 215, "times","multiplication sign, U+00D7 ISOnum" },
1853 { 216, "Oslash","latin capital letter O with stroke latin capital letter O slash, U+00D8 ISOlat1" },
1854 { 217, "Ugrave","latin capital letter U with grave, U+00D9 ISOlat1" },
1855 { 218, "Uacute","latin capital letter U with acute, U+00DA ISOlat1" },
1856 { 219, "Ucirc","latin capital letter U with circumflex, U+00DB ISOlat1" },
1857 { 220, "Uuml", "latin capital letter U with diaeresis, U+00DC ISOlat1" },
1858 { 221, "Yacute","latin capital letter Y with acute, U+00DD ISOlat1" },
1859 { 222, "THORN","latin capital letter THORN, U+00DE ISOlat1" },
1860 { 223, "szlig","latin small letter sharp s = ess-zed, U+00DF ISOlat1" },
1861 { 224, "agrave","latin small letter a with grave = latin small letter a grave, U+00E0 ISOlat1" },
1862 { 225, "aacute","latin small letter a with acute, U+00E1 ISOlat1" },
1863 { 226, "acirc","latin small letter a with circumflex, U+00E2 ISOlat1" },
1864 { 227, "atilde","latin small letter a with tilde, U+00E3 ISOlat1" },
1865 { 228, "auml", "latin small letter a with diaeresis, U+00E4 ISOlat1" },
1866 { 229, "aring","latin small letter a with ring above = latin small letter a ring, U+00E5 ISOlat1" },
1867 { 230, "aelig","latin small letter ae = latin small ligature ae, U+00E6 ISOlat1" },
1868 { 231, "ccedil","latin small letter c with cedilla, U+00E7 ISOlat1" },
1869 { 232, "egrave","latin small letter e with grave, U+00E8 ISOlat1" },
1870 { 233, "eacute","latin small letter e with acute, U+00E9 ISOlat1" },
1871 { 234, "ecirc","latin small letter e with circumflex, U+00EA ISOlat1" },
1872 { 235, "euml", "latin small letter e with diaeresis, U+00EB ISOlat1" },
1873 { 236, "igrave","latin small letter i with grave, U+00EC ISOlat1" },
1874 { 237, "iacute","latin small letter i with acute, U+00ED ISOlat1" },
1875 { 238, "icirc","latin small letter i with circumflex, U+00EE ISOlat1" },
1876 { 239, "iuml", "latin small letter i with diaeresis, U+00EF ISOlat1" },
1877 { 240, "eth", "latin small letter eth, U+00F0 ISOlat1" },
1878 { 241, "ntilde","latin small letter n with tilde, U+00F1 ISOlat1" },
1879 { 242, "ograve","latin small letter o with grave, U+00F2 ISOlat1" },
1880 { 243, "oacute","latin small letter o with acute, U+00F3 ISOlat1" },
1881 { 244, "ocirc","latin small letter o with circumflex, U+00F4 ISOlat1" },
1882 { 245, "otilde","latin small letter o with tilde, U+00F5 ISOlat1" },
1883 { 246, "ouml", "latin small letter o with diaeresis, U+00F6 ISOlat1" },
1884 { 247, "divide","division sign, U+00F7 ISOnum" },
1885 { 248, "oslash","latin small letter o with stroke, = latin small letter o slash, U+00F8 ISOlat1" },
1886 { 249, "ugrave","latin small letter u with grave, U+00F9 ISOlat1" },
1887 { 250, "uacute","latin small letter u with acute, U+00FA ISOlat1" },
1888 { 251, "ucirc","latin small letter u with circumflex, U+00FB ISOlat1" },
1889 { 252, "uuml", "latin small letter u with diaeresis, U+00FC ISOlat1" },
1890 { 253, "yacute","latin small letter y with acute, U+00FD ISOlat1" },
1891 { 254, "thorn","latin small letter thorn with, U+00FE ISOlat1" },
1892 { 255, "yuml", "latin small letter y with diaeresis, U+00FF ISOlat1" },
1894 { 338, "OElig","latin capital ligature OE, U+0152 ISOlat2" },
1895 { 339, "oelig","latin small ligature oe, U+0153 ISOlat2" },
1896 { 352, "Scaron","latin capital letter S with caron, U+0160 ISOlat2" },
1897 { 353, "scaron","latin small letter s with caron, U+0161 ISOlat2" },
1898 { 376, "Yuml", "latin capital letter Y with diaeresis, U+0178 ISOlat2" },
1901 * Anything below should really be kept as entities references
1903 { 402, "fnof", "latin small f with hook = function = florin, U+0192 ISOtech" },
1905 { 710, "circ", "modifier letter circumflex accent, U+02C6 ISOpub" },
1906 { 732, "tilde","small tilde, U+02DC ISOdia" },
1908 { 913, "Alpha","greek capital letter alpha, U+0391" },
1909 { 914, "Beta", "greek capital letter beta, U+0392" },
1910 { 915, "Gamma","greek capital letter gamma, U+0393 ISOgrk3" },
1911 { 916, "Delta","greek capital letter delta, U+0394 ISOgrk3" },
1912 { 917, "Epsilon","greek capital letter epsilon, U+0395" },
1913 { 918, "Zeta", "greek capital letter zeta, U+0396" },
1914 { 919, "Eta", "greek capital letter eta, U+0397" },
1915 { 920, "Theta","greek capital letter theta, U+0398 ISOgrk3" },
1916 { 921, "Iota", "greek capital letter iota, U+0399" },
1917 { 922, "Kappa","greek capital letter kappa, U+039A" },
1918 { 923, "Lambda", "greek capital letter lambda, U+039B ISOgrk3" },
1919 { 924, "Mu", "greek capital letter mu, U+039C" },
1920 { 925, "Nu", "greek capital letter nu, U+039D" },
1921 { 926, "Xi", "greek capital letter xi, U+039E ISOgrk3" },
1922 { 927, "Omicron","greek capital letter omicron, U+039F" },
1923 { 928, "Pi", "greek capital letter pi, U+03A0 ISOgrk3" },
1924 { 929, "Rho", "greek capital letter rho, U+03A1" },
1925 { 931, "Sigma","greek capital letter sigma, U+03A3 ISOgrk3" },
1926 { 932, "Tau", "greek capital letter tau, U+03A4" },
1927 { 933, "Upsilon","greek capital letter upsilon, U+03A5 ISOgrk3" },
1928 { 934, "Phi", "greek capital letter phi, U+03A6 ISOgrk3" },
1929 { 935, "Chi", "greek capital letter chi, U+03A7" },
1930 { 936, "Psi", "greek capital letter psi, U+03A8 ISOgrk3" },
1931 { 937, "Omega","greek capital letter omega, U+03A9 ISOgrk3" },
1933 { 945, "alpha","greek small letter alpha, U+03B1 ISOgrk3" },
1934 { 946, "beta", "greek small letter beta, U+03B2 ISOgrk3" },
1935 { 947, "gamma","greek small letter gamma, U+03B3 ISOgrk3" },
1936 { 948, "delta","greek small letter delta, U+03B4 ISOgrk3" },
1937 { 949, "epsilon","greek small letter epsilon, U+03B5 ISOgrk3" },
1938 { 950, "zeta", "greek small letter zeta, U+03B6 ISOgrk3" },
1939 { 951, "eta", "greek small letter eta, U+03B7 ISOgrk3" },
1940 { 952, "theta","greek small letter theta, U+03B8 ISOgrk3" },
1941 { 953, "iota", "greek small letter iota, U+03B9 ISOgrk3" },
1942 { 954, "kappa","greek small letter kappa, U+03BA ISOgrk3" },
1943 { 955, "lambda","greek small letter lambda, U+03BB ISOgrk3" },
1944 { 956, "mu", "greek small letter mu, U+03BC ISOgrk3" },
1945 { 957, "nu", "greek small letter nu, U+03BD ISOgrk3" },
1946 { 958, "xi", "greek small letter xi, U+03BE ISOgrk3" },
1947 { 959, "omicron","greek small letter omicron, U+03BF NEW" },
1948 { 960, "pi", "greek small letter pi, U+03C0 ISOgrk3" },
1949 { 961, "rho", "greek small letter rho, U+03C1 ISOgrk3" },
1950 { 962, "sigmaf","greek small letter final sigma, U+03C2 ISOgrk3" },
1951 { 963, "sigma","greek small letter sigma, U+03C3 ISOgrk3" },
1952 { 964, "tau", "greek small letter tau, U+03C4 ISOgrk3" },
1953 { 965, "upsilon","greek small letter upsilon, U+03C5 ISOgrk3" },
1954 { 966, "phi", "greek small letter phi, U+03C6 ISOgrk3" },
1955 { 967, "chi", "greek small letter chi, U+03C7 ISOgrk3" },
1956 { 968, "psi", "greek small letter psi, U+03C8 ISOgrk3" },
1957 { 969, "omega","greek small letter omega, U+03C9 ISOgrk3" },
1958 { 977, "thetasym","greek small letter theta symbol, U+03D1 NEW" },
1959 { 978, "upsih","greek upsilon with hook symbol, U+03D2 NEW" },
1960 { 982, "piv", "greek pi symbol, U+03D6 ISOgrk3" },
1962 { 8194, "ensp", "en space, U+2002 ISOpub" },
1963 { 8195, "emsp", "em space, U+2003 ISOpub" },
1964 { 8201, "thinsp","thin space, U+2009 ISOpub" },
1965 { 8204, "zwnj", "zero width non-joiner, U+200C NEW RFC 2070" },
1966 { 8205, "zwj", "zero width joiner, U+200D NEW RFC 2070" },
1967 { 8206, "lrm", "left-to-right mark, U+200E NEW RFC 2070" },
1968 { 8207, "rlm", "right-to-left mark, U+200F NEW RFC 2070" },
1969 { 8211, "ndash","en dash, U+2013 ISOpub" },
1970 { 8212, "mdash","em dash, U+2014 ISOpub" },
1971 { 8216, "lsquo","left single quotation mark, U+2018 ISOnum" },
1972 { 8217, "rsquo","right single quotation mark, U+2019 ISOnum" },
1973 { 8218, "sbquo","single low-9 quotation mark, U+201A NEW" },
1974 { 8220, "ldquo","left double quotation mark, U+201C ISOnum" },
1975 { 8221, "rdquo","right double quotation mark, U+201D ISOnum" },
1976 { 8222, "bdquo","double low-9 quotation mark, U+201E NEW" },
1977 { 8224, "dagger","dagger, U+2020 ISOpub" },
1978 { 8225, "Dagger","double dagger, U+2021 ISOpub" },
1980 { 8226, "bull", "bullet = black small circle, U+2022 ISOpub" },
1981 { 8230, "hellip","horizontal ellipsis = three dot leader, U+2026 ISOpub" },
1983 { 8240, "permil","per mille sign, U+2030 ISOtech" },
1985 { 8242, "prime","prime = minutes = feet, U+2032 ISOtech" },
1986 { 8243, "Prime","double prime = seconds = inches, U+2033 ISOtech" },
1988 { 8249, "lsaquo","single left-pointing angle quotation mark, U+2039 ISO proposed" },
1989 { 8250, "rsaquo","single right-pointing angle quotation mark, U+203A ISO proposed" },
1991 { 8254, "oline","overline = spacing overscore, U+203E NEW" },
1992 { 8260, "frasl","fraction slash, U+2044 NEW" },
1994 { 8364, "euro", "euro sign, U+20AC NEW" },
1996 { 8465, "image","blackletter capital I = imaginary part, U+2111 ISOamso" },
1997 { 8472, "weierp","script capital P = power set = Weierstrass p, U+2118 ISOamso" },
1998 { 8476, "real", "blackletter capital R = real part symbol, U+211C ISOamso" },
1999 { 8482, "trade","trade mark sign, U+2122 ISOnum" },
2000 { 8501, "alefsym","alef symbol = first transfinite cardinal, U+2135 NEW" },
2001 { 8592, "larr", "leftwards arrow, U+2190 ISOnum" },
2002 { 8593, "uarr", "upwards arrow, U+2191 ISOnum" },
2003 { 8594, "rarr", "rightwards arrow, U+2192 ISOnum" },
2004 { 8595, "darr", "downwards arrow, U+2193 ISOnum" },
2005 { 8596, "harr", "left right arrow, U+2194 ISOamsa" },
2006 { 8629, "crarr","downwards arrow with corner leftwards = carriage return, U+21B5 NEW" },
2007 { 8656, "lArr", "leftwards double arrow, U+21D0 ISOtech" },
2008 { 8657, "uArr", "upwards double arrow, U+21D1 ISOamsa" },
2009 { 8658, "rArr", "rightwards double arrow, U+21D2 ISOtech" },
2010 { 8659, "dArr", "downwards double arrow, U+21D3 ISOamsa" },
2011 { 8660, "hArr", "left right double arrow, U+21D4 ISOamsa" },
2013 { 8704, "forall","for all, U+2200 ISOtech" },
2014 { 8706, "part", "partial differential, U+2202 ISOtech" },
2015 { 8707, "exist","there exists, U+2203 ISOtech" },
2016 { 8709, "empty","empty set = null set = diameter, U+2205 ISOamso" },
2017 { 8711, "nabla","nabla = backward difference, U+2207 ISOtech" },
2018 { 8712, "isin", "element of, U+2208 ISOtech" },
2019 { 8713, "notin","not an element of, U+2209 ISOtech" },
2020 { 8715, "ni", "contains as member, U+220B ISOtech" },
2021 { 8719, "prod", "n-ary product = product sign, U+220F ISOamsb" },
2022 { 8721, "sum", "n-ary summation, U+2211 ISOamsb" },
2023 { 8722, "minus","minus sign, U+2212 ISOtech" },
2024 { 8727, "lowast","asterisk operator, U+2217 ISOtech" },
2025 { 8730, "radic","square root = radical sign, U+221A ISOtech" },
2026 { 8733, "prop", "proportional to, U+221D ISOtech" },
2027 { 8734, "infin","infinity, U+221E ISOtech" },
2028 { 8736, "ang", "angle, U+2220 ISOamso" },
2029 { 8743, "and", "logical and = wedge, U+2227 ISOtech" },
2030 { 8744, "or", "logical or = vee, U+2228 ISOtech" },
2031 { 8745, "cap", "intersection = cap, U+2229 ISOtech" },
2032 { 8746, "cup", "union = cup, U+222A ISOtech" },
2033 { 8747, "int", "integral, U+222B ISOtech" },
2034 { 8756, "there4","therefore, U+2234 ISOtech" },
2035 { 8764, "sim", "tilde operator = varies with = similar to, U+223C ISOtech" },
2036 { 8773, "cong", "approximately equal to, U+2245 ISOtech" },
2037 { 8776, "asymp","almost equal to = asymptotic to, U+2248 ISOamsr" },
2038 { 8800, "ne", "not equal to, U+2260 ISOtech" },
2039 { 8801, "equiv","identical to, U+2261 ISOtech" },
2040 { 8804, "le", "less-than or equal to, U+2264 ISOtech" },
2041 { 8805, "ge", "greater-than or equal to, U+2265 ISOtech" },
2042 { 8834, "sub", "subset of, U+2282 ISOtech" },
2043 { 8835, "sup", "superset of, U+2283 ISOtech" },
2044 { 8836, "nsub", "not a subset of, U+2284 ISOamsn" },
2045 { 8838, "sube", "subset of or equal to, U+2286 ISOtech" },
2046 { 8839, "supe", "superset of or equal to, U+2287 ISOtech" },
2047 { 8853, "oplus","circled plus = direct sum, U+2295 ISOamsb" },
2048 { 8855, "otimes","circled times = vector product, U+2297 ISOamsb" },
2049 { 8869, "perp", "up tack = orthogonal to = perpendicular, U+22A5 ISOtech" },
2050 { 8901, "sdot", "dot operator, U+22C5 ISOamsb" },
2051 { 8968, "lceil","left ceiling = apl upstile, U+2308 ISOamsc" },
2052 { 8969, "rceil","right ceiling, U+2309 ISOamsc" },
2053 { 8970, "lfloor","left floor = apl downstile, U+230A ISOamsc" },
2054 { 8971, "rfloor","right floor, U+230B ISOamsc" },
2055 { 9001, "lang", "left-pointing angle bracket = bra, U+2329 ISOtech" },
2056 { 9002, "rang", "right-pointing angle bracket = ket, U+232A ISOtech" },
2057 { 9674, "loz", "lozenge, U+25CA ISOpub" },
2059 { 9824, "spades","black spade suit, U+2660 ISOpub" },
2060 { 9827, "clubs","black club suit = shamrock, U+2663 ISOpub" },
2061 { 9829, "hearts","black heart suit = valentine, U+2665 ISOpub" },
2062 { 9830, "diams","black diamond suit, U+2666 ISOpub" },
2066 /************************************************************************
2068 * Commodity functions to handle entities *
2070 ************************************************************************/
2073 * Macro used to grow the current buffer.
2075 #define growBuffer(buffer) { \
2077 buffer##_size *= 2; \
2078 tmp = (xmlChar *) xmlRealloc(buffer, buffer##_size * sizeof(xmlChar)); \
2079 if (tmp == NULL) { \
2080 htmlErrMemory(ctxt, "growing buffer\n"); \
2089 * @name: the entity name
2091 * Lookup the given entity in EntitiesTable
2093 * TODO: the linear scan is really ugly, an hash table is really needed.
2095 * Returns the associated htmlEntityDescPtr if found, NULL otherwise.
2097 const htmlEntityDesc
*
2098 htmlEntityLookup(const xmlChar
*name
) {
2101 for (i
= 0;i
< (sizeof(html40EntitiesTable
)/
2102 sizeof(html40EntitiesTable
[0]));i
++) {
2103 if (xmlStrEqual(name
, BAD_CAST html40EntitiesTable
[i
].name
)) {
2104 return((htmlEntityDescPtr
) &html40EntitiesTable
[i
]);
2111 * htmlEntityValueLookup:
2112 * @value: the entity's unicode value
2114 * Lookup the given entity in EntitiesTable
2116 * TODO: the linear scan is really ugly, an hash table is really needed.
2118 * Returns the associated htmlEntityDescPtr if found, NULL otherwise.
2120 const htmlEntityDesc
*
2121 htmlEntityValueLookup(unsigned int value
) {
2124 for (i
= 0;i
< (sizeof(html40EntitiesTable
)/
2125 sizeof(html40EntitiesTable
[0]));i
++) {
2126 if (html40EntitiesTable
[i
].value
>= value
) {
2127 if (html40EntitiesTable
[i
].value
> value
)
2129 return((htmlEntityDescPtr
) &html40EntitiesTable
[i
]);
2137 * @out: a pointer to an array of bytes to store the result
2138 * @outlen: the length of @out
2139 * @in: a pointer to an array of UTF-8 chars
2140 * @inlen: the length of @in
2142 * Take a block of UTF-8 chars in and try to convert it to an ASCII
2143 * plus HTML entities block of chars out.
2145 * Returns 0 if success, -2 if the transcoding fails, or -1 otherwise
2146 * The value of @inlen after return is the number of octets consumed
2147 * as the return value is positive, else unpredictable.
2148 * The value of @outlen after return is the number of octets consumed.
2151 UTF8ToHtml(unsigned char* out
, int *outlen
,
2152 const unsigned char* in
, int *inlen
) {
2153 const unsigned char* processed
= in
;
2154 const unsigned char* outend
;
2155 const unsigned char* outstart
= out
;
2156 const unsigned char* instart
= in
;
2157 const unsigned char* inend
;
2161 if ((out
== NULL
) || (outlen
== NULL
) || (inlen
== NULL
)) return(-1);
2164 * initialization nothing to do
2170 inend
= in
+ (*inlen
);
2171 outend
= out
+ (*outlen
);
2172 while (in
< inend
) {
2174 if (d
< 0x80) { c
= d
; trailing
= 0; }
2175 else if (d
< 0xC0) {
2176 /* trailing byte in leading position */
2177 *outlen
= out
- outstart
;
2178 *inlen
= processed
- instart
;
2180 } else if (d
< 0xE0) { c
= d
& 0x1F; trailing
= 1; }
2181 else if (d
< 0xF0) { c
= d
& 0x0F; trailing
= 2; }
2182 else if (d
< 0xF8) { c
= d
& 0x07; trailing
= 3; }
2184 /* no chance for this in Ascii */
2185 *outlen
= out
- outstart
;
2186 *inlen
= processed
- instart
;
2190 if (inend
- in
< trailing
) {
2194 for ( ; trailing
; trailing
--) {
2195 if ((in
>= inend
) || (((d
= *in
++) & 0xC0) != 0x80))
2201 /* assertion: c is a single UTF-4 value */
2203 if (out
+ 1 >= outend
)
2208 const htmlEntityDesc
* ent
;
2213 * Try to lookup a predefined HTML entity for it
2216 ent
= htmlEntityValueLookup(c
);
2218 snprintf(nbuf
, sizeof(nbuf
), "#%u", c
);
2224 if (out
+ 2 + len
>= outend
)
2227 memcpy(out
, cp
, len
);
2233 *outlen
= out
- outstart
;
2234 *inlen
= processed
- instart
;
2239 * htmlEncodeEntities:
2240 * @out: a pointer to an array of bytes to store the result
2241 * @outlen: the length of @out
2242 * @in: a pointer to an array of UTF-8 chars
2243 * @inlen: the length of @in
2244 * @quoteChar: the quote character to escape (' or ") or zero.
2246 * Take a block of UTF-8 chars in and try to convert it to an ASCII
2247 * plus HTML entities block of chars out.
2249 * Returns 0 if success, -2 if the transcoding fails, or -1 otherwise
2250 * The value of @inlen after return is the number of octets consumed
2251 * as the return value is positive, else unpredictable.
2252 * The value of @outlen after return is the number of octets consumed.
2255 htmlEncodeEntities(unsigned char* out
, int *outlen
,
2256 const unsigned char* in
, int *inlen
, int quoteChar
) {
2257 const unsigned char* processed
= in
;
2258 const unsigned char* outend
;
2259 const unsigned char* outstart
= out
;
2260 const unsigned char* instart
= in
;
2261 const unsigned char* inend
;
2265 if ((out
== NULL
) || (outlen
== NULL
) || (inlen
== NULL
) || (in
== NULL
))
2267 outend
= out
+ (*outlen
);
2268 inend
= in
+ (*inlen
);
2269 while (in
< inend
) {
2271 if (d
< 0x80) { c
= d
; trailing
= 0; }
2272 else if (d
< 0xC0) {
2273 /* trailing byte in leading position */
2274 *outlen
= out
- outstart
;
2275 *inlen
= processed
- instart
;
2277 } else if (d
< 0xE0) { c
= d
& 0x1F; trailing
= 1; }
2278 else if (d
< 0xF0) { c
= d
& 0x0F; trailing
= 2; }
2279 else if (d
< 0xF8) { c
= d
& 0x07; trailing
= 3; }
2281 /* no chance for this in Ascii */
2282 *outlen
= out
- outstart
;
2283 *inlen
= processed
- instart
;
2287 if (inend
- in
< trailing
)
2290 while (trailing
--) {
2291 if (((d
= *in
++) & 0xC0) != 0x80) {
2292 *outlen
= out
- outstart
;
2293 *inlen
= processed
- instart
;
2300 /* assertion: c is a single UTF-4 value */
2301 if ((c
< 0x80) && (c
!= (unsigned int) quoteChar
) &&
2302 (c
!= '&') && (c
!= '<') && (c
!= '>')) {
2307 const htmlEntityDesc
* ent
;
2313 * Try to lookup a predefined HTML entity for it
2315 ent
= htmlEntityValueLookup(c
);
2317 snprintf(nbuf
, sizeof(nbuf
), "#%u", c
);
2323 if (out
+ 2 + len
> outend
)
2326 memcpy(out
, cp
, len
);
2332 *outlen
= out
- outstart
;
2333 *inlen
= processed
- instart
;
2337 /************************************************************************
2339 * Commodity functions to handle streams *
2341 ************************************************************************/
2343 #ifdef LIBXML_PUSH_ENABLED
2345 * htmlNewInputStream:
2346 * @ctxt: an HTML parser context
2348 * Create a new input stream structure
2349 * Returns the new input stream or NULL
2351 static htmlParserInputPtr
2352 htmlNewInputStream(htmlParserCtxtPtr ctxt
) {
2353 htmlParserInputPtr input
;
2355 input
= (xmlParserInputPtr
) xmlMalloc(sizeof(htmlParserInput
));
2356 if (input
== NULL
) {
2357 htmlErrMemory(ctxt
, "couldn't allocate a new input stream\n");
2360 memset(input
, 0, sizeof(htmlParserInput
));
2361 input
->filename
= NULL
;
2362 input
->directory
= NULL
;
2370 input
->version
= NULL
;
2371 input
->consumed
= 0;
2378 /************************************************************************
2380 * Commodity functions, cleanup needed ? *
2382 ************************************************************************/
2384 * all tags allowing pc data from the html 4.01 loose dtd
2385 * NOTE: it might be more appropriate to integrate this information
2386 * into the html40ElementTable array but I don't want to risk any
2387 * binary incompatibility
2389 static const char *allowPCData
[] = {
2390 "a", "abbr", "acronym", "address", "applet", "b", "bdo", "big",
2391 "blockquote", "body", "button", "caption", "center", "cite", "code",
2392 "dd", "del", "dfn", "div", "dt", "em", "font", "form", "h1", "h2",
2393 "h3", "h4", "h5", "h6", "i", "iframe", "ins", "kbd", "label", "legend",
2394 "li", "noframes", "noscript", "object", "p", "pre", "q", "s", "samp",
2395 "small", "span", "strike", "strong", "td", "th", "tt", "u", "var"
2400 * @ctxt: an HTML parser context
2402 * @len: the size of @str
2404 * Is this a sequence of blank chars that one can ignore ?
2406 * Returns 1 if ignorable 0 otherwise.
2409 static int areBlanks(htmlParserCtxtPtr ctxt
, const xmlChar
*str
, int len
) {
2412 xmlNodePtr lastChild
;
2415 for (j
= 0;j
< len
;j
++)
2416 if (!(IS_BLANK_CH(str
[j
]))) return(0);
2418 if (CUR
== 0) return(1);
2419 if (CUR
!= '<') return(0);
2420 if (ctxt
->name
== NULL
)
2422 if (xmlStrEqual(ctxt
->name
, BAD_CAST
"html"))
2424 if (xmlStrEqual(ctxt
->name
, BAD_CAST
"head"))
2427 /* Only strip CDATA children of the body tag for strict HTML DTDs */
2428 if (xmlStrEqual(ctxt
->name
, BAD_CAST
"body") && ctxt
->myDoc
!= NULL
) {
2429 dtd
= xmlGetIntSubset(ctxt
->myDoc
);
2430 if (dtd
!= NULL
&& dtd
->ExternalID
!= NULL
) {
2431 if (!xmlStrcasecmp(dtd
->ExternalID
, BAD_CAST
"-//W3C//DTD HTML 4.01//EN") ||
2432 !xmlStrcasecmp(dtd
->ExternalID
, BAD_CAST
"-//W3C//DTD HTML 4//EN"))
2437 if (ctxt
->node
== NULL
) return(0);
2438 lastChild
= xmlGetLastChild(ctxt
->node
);
2439 while ((lastChild
) && (lastChild
->type
== XML_COMMENT_NODE
))
2440 lastChild
= lastChild
->prev
;
2441 if (lastChild
== NULL
) {
2442 if ((ctxt
->node
->type
!= XML_ELEMENT_NODE
) &&
2443 (ctxt
->node
->content
!= NULL
)) return(0);
2444 /* keep ws in constructs like ...<b> </b>...
2445 for all tags "b" allowing PCDATA */
2446 for ( i
= 0; i
< sizeof(allowPCData
)/sizeof(allowPCData
[0]); i
++ ) {
2447 if ( xmlStrEqual(ctxt
->name
, BAD_CAST allowPCData
[i
]) ) {
2451 } else if (xmlNodeIsText(lastChild
)) {
2454 /* keep ws in constructs like <p><b>xy</b> <i>z</i><p>
2455 for all tags "p" allowing PCDATA */
2456 for ( i
= 0; i
< sizeof(allowPCData
)/sizeof(allowPCData
[0]); i
++ ) {
2457 if ( xmlStrEqual(lastChild
->name
, BAD_CAST allowPCData
[i
]) ) {
2467 * @URI: URI for the dtd, or NULL
2468 * @ExternalID: the external ID of the DTD, or NULL
2470 * Creates a new HTML document without a DTD node if @URI and @ExternalID
2473 * Returns a new document, do not initialize the DTD if not provided
2476 htmlNewDocNoDtD(const xmlChar
*URI
, const xmlChar
*ExternalID
) {
2480 * Allocate a new document and fill the fields.
2482 cur
= (xmlDocPtr
) xmlMalloc(sizeof(xmlDoc
));
2484 htmlErrMemory(NULL
, "HTML document creation failed\n");
2487 memset(cur
, 0, sizeof(xmlDoc
));
2489 cur
->type
= XML_HTML_DOCUMENT_NODE
;
2490 cur
->version
= NULL
;
2491 cur
->intSubset
= NULL
;
2494 cur
->children
= NULL
;
2495 cur
->extSubset
= NULL
;
2497 cur
->encoding
= NULL
;
2498 cur
->standalone
= 1;
2499 cur
->compression
= 0;
2502 cur
->_private
= NULL
;
2503 cur
->charset
= XML_CHAR_ENCODING_UTF8
;
2504 cur
->properties
= XML_DOC_HTML
| XML_DOC_USERBUILT
;
2505 if ((ExternalID
!= NULL
) ||
2507 xmlCreateIntSubset(cur
, BAD_CAST
"html", ExternalID
, URI
);
2508 if ((__xmlRegisterCallbacks
) && (xmlRegisterNodeDefaultValue
))
2509 xmlRegisterNodeDefaultValue((xmlNodePtr
)cur
);
2515 * @URI: URI for the dtd, or NULL
2516 * @ExternalID: the external ID of the DTD, or NULL
2518 * Creates a new HTML document
2520 * Returns a new document
2523 htmlNewDoc(const xmlChar
*URI
, const xmlChar
*ExternalID
) {
2524 if ((URI
== NULL
) && (ExternalID
== NULL
))
2525 return(htmlNewDocNoDtD(
2526 BAD_CAST
"http://www.w3.org/TR/REC-html40/loose.dtd",
2527 BAD_CAST
"-//W3C//DTD HTML 4.0 Transitional//EN"));
2529 return(htmlNewDocNoDtD(URI
, ExternalID
));
2533 /************************************************************************
2535 * The parser itself *
2536 * Relates to http://www.w3.org/TR/html40 *
2538 ************************************************************************/
2540 /************************************************************************
2542 * The parser itself *
2544 ************************************************************************/
2546 static const xmlChar
* htmlParseNameComplex(xmlParserCtxtPtr ctxt
);
2549 htmlSkipBogusComment(htmlParserCtxtPtr ctxt
) {
2552 htmlParseErr(ctxt
, XML_HTML_INCORRECTLY_OPENED_COMMENT
,
2553 "Incorrectly opened comment\n", NULL
, NULL
);
2564 * htmlParseHTMLName:
2565 * @ctxt: an HTML parser context
2567 * parse an HTML tag or attribute name, note that we convert it to lowercase
2568 * since HTML names are not case-sensitive.
2570 * Returns the Tag Name parsed or NULL
2573 static const xmlChar
*
2574 htmlParseHTMLName(htmlParserCtxtPtr ctxt
) {
2576 xmlChar loc
[HTML_PARSER_BUFFER_SIZE
];
2578 if (!IS_ASCII_LETTER(CUR
) && (CUR
!= '_') &&
2579 (CUR
!= ':') && (CUR
!= '.')) return(NULL
);
2581 while ((i
< HTML_PARSER_BUFFER_SIZE
) &&
2582 ((IS_ASCII_LETTER(CUR
)) || (IS_ASCII_DIGIT(CUR
)) ||
2583 (CUR
== ':') || (CUR
== '-') || (CUR
== '_') ||
2585 if ((CUR
>= 'A') && (CUR
<= 'Z')) loc
[i
] = CUR
+ 0x20;
2592 return(xmlDictLookup(ctxt
->dict
, loc
, i
));
2597 * htmlParseHTMLName_nonInvasive:
2598 * @ctxt: an HTML parser context
2600 * parse an HTML tag or attribute name, note that we convert it to lowercase
2601 * since HTML names are not case-sensitive, this doesn't consume the data
2602 * from the stream, it's a look-ahead
2604 * Returns the Tag Name parsed or NULL
2607 static const xmlChar
*
2608 htmlParseHTMLName_nonInvasive(htmlParserCtxtPtr ctxt
) {
2610 xmlChar loc
[HTML_PARSER_BUFFER_SIZE
];
2612 if (!IS_ASCII_LETTER(NXT(1)) && (NXT(1) != '_') &&
2613 (NXT(1) != ':')) return(NULL
);
2615 while ((i
< HTML_PARSER_BUFFER_SIZE
) &&
2616 ((IS_ASCII_LETTER(NXT(1+i
))) || (IS_ASCII_DIGIT(NXT(1+i
))) ||
2617 (NXT(1+i
) == ':') || (NXT(1+i
) == '-') || (NXT(1+i
) == '_'))) {
2618 if ((NXT(1+i
) >= 'A') && (NXT(1+i
) <= 'Z')) loc
[i
] = NXT(1+i
) + 0x20;
2619 else loc
[i
] = NXT(1+i
);
2623 return(xmlDictLookup(ctxt
->dict
, loc
, i
));
2629 * @ctxt: an HTML parser context
2631 * parse an HTML name, this routine is case sensitive.
2633 * Returns the Name parsed or NULL
2636 static const xmlChar
*
2637 htmlParseName(htmlParserCtxtPtr ctxt
) {
2645 * Accelerator for simple ASCII names
2647 in
= ctxt
->input
->cur
;
2648 if (((*in
>= 0x61) && (*in
<= 0x7A)) ||
2649 ((*in
>= 0x41) && (*in
<= 0x5A)) ||
2650 (*in
== '_') || (*in
== ':')) {
2652 while (((*in
>= 0x61) && (*in
<= 0x7A)) ||
2653 ((*in
>= 0x41) && (*in
<= 0x5A)) ||
2654 ((*in
>= 0x30) && (*in
<= 0x39)) ||
2655 (*in
== '_') || (*in
== '-') ||
2656 (*in
== ':') || (*in
== '.'))
2659 if (in
== ctxt
->input
->end
)
2662 if ((*in
> 0) && (*in
< 0x80)) {
2663 count
= in
- ctxt
->input
->cur
;
2664 ret
= xmlDictLookup(ctxt
->dict
, ctxt
->input
->cur
, count
);
2665 ctxt
->input
->cur
= in
;
2666 ctxt
->input
->col
+= count
;
2670 return(htmlParseNameComplex(ctxt
));
2673 static const xmlChar
*
2674 htmlParseNameComplex(xmlParserCtxtPtr ctxt
) {
2678 const xmlChar
*base
= ctxt
->input
->base
;
2681 * Handler for more complex cases
2685 if ((c
== ' ') || (c
== '>') || (c
== '/') || /* accelerators */
2686 (!IS_LETTER(c
) && (c
!= '_') &&
2691 while ((c
!= ' ') && (c
!= '>') && (c
!= '/') && /* test bigname.xml */
2692 ((IS_LETTER(c
)) || (IS_DIGIT(c
)) ||
2693 (c
== '.') || (c
== '-') ||
2694 (c
== '_') || (c
== ':') ||
2695 (IS_COMBINING(c
)) ||
2696 (IS_EXTENDER(c
)))) {
2697 if (count
++ > 100) {
2704 if (ctxt
->input
->base
!= base
) {
2706 * We changed encoding from an unknown encoding
2707 * Input buffer changed location, so we better start again
2709 return(htmlParseNameComplex(ctxt
));
2713 if (ctxt
->input
->cur
- ctxt
->input
->base
< len
) {
2715 htmlParseErr(ctxt
, XML_ERR_INTERNAL_ERROR
,
2716 "unexpected change of input buffer", NULL
, NULL
);
2720 return(xmlDictLookup(ctxt
->dict
, ctxt
->input
->cur
- len
, len
));
2725 * htmlParseHTMLAttribute:
2726 * @ctxt: an HTML parser context
2727 * @stop: a char stop value
2729 * parse an HTML attribute value till the stop (quote), if
2730 * stop is 0 then it stops at the first space
2732 * Returns the attribute parsed or NULL
2736 htmlParseHTMLAttribute(htmlParserCtxtPtr ctxt
, const xmlChar stop
) {
2737 xmlChar
*buffer
= NULL
;
2738 int buffer_size
= 0;
2739 xmlChar
*out
= NULL
;
2740 const xmlChar
*name
= NULL
;
2741 const xmlChar
*cur
= NULL
;
2742 const htmlEntityDesc
* ent
;
2745 * allocate a translation buffer.
2747 buffer_size
= HTML_PARSER_BUFFER_SIZE
;
2748 buffer
= (xmlChar
*) xmlMallocAtomic(buffer_size
* sizeof(xmlChar
));
2749 if (buffer
== NULL
) {
2750 htmlErrMemory(ctxt
, "buffer allocation failed\n");
2756 * Ok loop until we reach one of the ending chars
2758 while ((CUR
!= 0) && (CUR
!= stop
)) {
2759 if ((stop
== 0) && (CUR
== '>')) break;
2760 if ((stop
== 0) && (IS_BLANK_CH(CUR
))) break;
2762 if (NXT(1) == '#') {
2766 c
= htmlParseCharRef(ctxt
);
2768 { *out
++ = c
; bits
= -6; }
2770 { *out
++ =((c
>> 6) & 0x1F) | 0xC0; bits
= 0; }
2771 else if (c
< 0x10000)
2772 { *out
++ =((c
>> 12) & 0x0F) | 0xE0; bits
= 6; }
2774 { *out
++ =((c
>> 18) & 0x07) | 0xF0; bits
= 12; }
2776 for ( ; bits
>= 0; bits
-= 6) {
2777 *out
++ = ((c
>> bits
) & 0x3F) | 0x80;
2780 if (out
- buffer
> buffer_size
- 100) {
2781 int indx
= out
- buffer
;
2784 out
= &buffer
[indx
];
2787 ent
= htmlParseEntityRef(ctxt
, &name
);
2790 if (out
- buffer
> buffer_size
- 100) {
2791 int indx
= out
- buffer
;
2794 out
= &buffer
[indx
];
2796 } else if (ent
== NULL
) {
2800 if (out
- buffer
> buffer_size
- 100) {
2801 int indx
= out
- buffer
;
2804 out
= &buffer
[indx
];
2812 if (out
- buffer
> buffer_size
- 100) {
2813 int indx
= out
- buffer
;
2816 out
= &buffer
[indx
];
2820 { *out
++ = c
; bits
= -6; }
2822 { *out
++ =((c
>> 6) & 0x1F) | 0xC0; bits
= 0; }
2823 else if (c
< 0x10000)
2824 { *out
++ =((c
>> 12) & 0x0F) | 0xE0; bits
= 6; }
2826 { *out
++ =((c
>> 18) & 0x07) | 0xF0; bits
= 12; }
2828 for ( ; bits
>= 0; bits
-= 6) {
2829 *out
++ = ((c
>> bits
) & 0x3F) | 0x80;
2837 if (out
- buffer
> buffer_size
- 100) {
2838 int indx
= out
- buffer
;
2841 out
= &buffer
[indx
];
2845 { *out
++ = c
; bits
= -6; }
2847 { *out
++ =((c
>> 6) & 0x1F) | 0xC0; bits
= 0; }
2848 else if (c
< 0x10000)
2849 { *out
++ =((c
>> 12) & 0x0F) | 0xE0; bits
= 6; }
2851 { *out
++ =((c
>> 18) & 0x07) | 0xF0; bits
= 12; }
2853 for ( ; bits
>= 0; bits
-= 6) {
2854 *out
++ = ((c
>> bits
) & 0x3F) | 0x80;
2864 * htmlParseEntityRef:
2865 * @ctxt: an HTML parser context
2866 * @str: location to store the entity name
2868 * parse an HTML ENTITY references
2870 * [68] EntityRef ::= '&' Name ';'
2872 * Returns the associated htmlEntityDescPtr if found, or NULL otherwise,
2873 * if non-NULL *str will have to be freed by the caller.
2875 const htmlEntityDesc
*
2876 htmlParseEntityRef(htmlParserCtxtPtr ctxt
, const xmlChar
**str
) {
2877 const xmlChar
*name
;
2878 const htmlEntityDesc
* ent
= NULL
;
2880 if (str
!= NULL
) *str
= NULL
;
2881 if ((ctxt
== NULL
) || (ctxt
->input
== NULL
)) return(NULL
);
2885 name
= htmlParseName(ctxt
);
2887 htmlParseErr(ctxt
, XML_ERR_NAME_REQUIRED
,
2888 "htmlParseEntityRef: no name\n", NULL
, NULL
);
2896 * Lookup the entity in the table.
2898 ent
= htmlEntityLookup(name
);
2899 if (ent
!= NULL
) /* OK that's ugly !!! */
2902 htmlParseErr(ctxt
, XML_ERR_ENTITYREF_SEMICOL_MISSING
,
2903 "htmlParseEntityRef: expecting ';'\n",
2914 * htmlParseAttValue:
2915 * @ctxt: an HTML parser context
2917 * parse a value for an attribute
2918 * Note: the parser won't do substitution of entities here, this
2919 * will be handled later in xmlStringGetNodeList, unless it was
2920 * asked for ctxt->replaceEntities != 0
2922 * Returns the AttValue parsed or NULL.
2926 htmlParseAttValue(htmlParserCtxtPtr ctxt
) {
2927 xmlChar
*ret
= NULL
;
2931 ret
= htmlParseHTMLAttribute(ctxt
, '"');
2933 htmlParseErr(ctxt
, XML_ERR_ATTRIBUTE_NOT_FINISHED
,
2934 "AttValue: \" expected\n", NULL
, NULL
);
2937 } else if (CUR
== '\'') {
2939 ret
= htmlParseHTMLAttribute(ctxt
, '\'');
2941 htmlParseErr(ctxt
, XML_ERR_ATTRIBUTE_NOT_FINISHED
,
2942 "AttValue: ' expected\n", NULL
, NULL
);
2947 * That's an HTMLism, the attribute value may not be quoted
2949 ret
= htmlParseHTMLAttribute(ctxt
, 0);
2951 htmlParseErr(ctxt
, XML_ERR_ATTRIBUTE_WITHOUT_VALUE
,
2952 "AttValue: no value found\n", NULL
, NULL
);
2959 * htmlParseSystemLiteral:
2960 * @ctxt: an HTML parser context
2962 * parse an HTML Literal
2964 * [11] SystemLiteral ::= ('"' [^"]* '"') | ("'" [^']* "'")
2966 * Returns the SystemLiteral parsed or NULL
2970 htmlParseSystemLiteral(htmlParserCtxtPtr ctxt
) {
2971 size_t len
= 0, startPosition
= 0;
2974 xmlChar
*ret
= NULL
;
2976 if ((CUR
!= '"') && (CUR
!= '\'')) {
2977 htmlParseErr(ctxt
, XML_ERR_LITERAL_NOT_STARTED
,
2978 "SystemLiteral \" or ' expected\n", NULL
, NULL
);
2984 if (CUR_PTR
< BASE_PTR
)
2986 startPosition
= CUR_PTR
- BASE_PTR
;
2988 while ((CUR
!= 0) && (CUR
!= quote
)) {
2989 /* TODO: Handle UTF-8 */
2990 if (!IS_CHAR_CH(CUR
)) {
2991 htmlParseErrInt(ctxt
, XML_ERR_INVALID_CHAR
,
2992 "Invalid char in SystemLiteral 0x%X\n", CUR
);
2999 htmlParseErr(ctxt
, XML_ERR_LITERAL_NOT_FINISHED
,
3000 "Unfinished SystemLiteral\n", NULL
, NULL
);
3004 ret
= xmlStrndup((BASE_PTR
+startPosition
), len
);
3011 * htmlParsePubidLiteral:
3012 * @ctxt: an HTML parser context
3014 * parse an HTML public literal
3016 * [12] PubidLiteral ::= '"' PubidChar* '"' | "'" (PubidChar - "'")* "'"
3018 * Returns the PubidLiteral parsed or NULL.
3022 htmlParsePubidLiteral(htmlParserCtxtPtr ctxt
) {
3023 size_t len
= 0, startPosition
= 0;
3026 xmlChar
*ret
= NULL
;
3028 if ((CUR
!= '"') && (CUR
!= '\'')) {
3029 htmlParseErr(ctxt
, XML_ERR_LITERAL_NOT_STARTED
,
3030 "PubidLiteral \" or ' expected\n", NULL
, NULL
);
3037 * Name ::= (Letter | '_') (NameChar)*
3039 if (CUR_PTR
< BASE_PTR
)
3041 startPosition
= CUR_PTR
- BASE_PTR
;
3043 while ((CUR
!= 0) && (CUR
!= quote
)) {
3044 if (!IS_PUBIDCHAR_CH(CUR
)) {
3045 htmlParseErrInt(ctxt
, XML_ERR_INVALID_CHAR
,
3046 "Invalid char in PubidLiteral 0x%X\n", CUR
);
3054 htmlParseErr(ctxt
, XML_ERR_LITERAL_NOT_FINISHED
,
3055 "Unfinished PubidLiteral\n", NULL
, NULL
);
3059 ret
= xmlStrndup((BASE_PTR
+ startPosition
), len
);
3067 * @ctxt: an HTML parser context
3069 * parse the content of an HTML SCRIPT or STYLE element
3070 * http://www.w3.org/TR/html4/sgml/dtd.html#Script
3071 * http://www.w3.org/TR/html4/sgml/dtd.html#StyleSheet
3072 * http://www.w3.org/TR/html4/types.html#type-script
3073 * http://www.w3.org/TR/html4/types.html#h-6.15
3074 * http://www.w3.org/TR/html4/appendix/notes.html#h-B.3.2.1
3076 * Script data ( %Script; in the DTD) can be the content of the SCRIPT
3077 * element and the value of intrinsic event attributes. User agents must
3078 * not evaluate script data as HTML markup but instead must pass it on as
3079 * data to a script engine.
3081 * - The content is passed like CDATA
3082 * - the attributes for style and scripting "onXXX" are also described
3083 * as CDATA but SGML allows entities references in attributes so their
3084 * processing is identical as other attributes
3087 htmlParseScript(htmlParserCtxtPtr ctxt
) {
3088 xmlChar buf
[HTML_PARSER_BIG_BUFFER_SIZE
+ 5];
3095 if ((cur
== '<') && (NXT(1) == '/')) {
3097 * One should break here, the specification is clear:
3098 * Authors should therefore escape "</" within the content.
3099 * Escape mechanisms are specific to each scripting or
3100 * style sheet language.
3102 * In recovery mode, only break if end tag match the
3103 * current tag, effectively ignoring all tags inside the
3104 * script/style block and treating the entire block as
3107 if (ctxt
->recovery
) {
3108 if (xmlStrncasecmp(ctxt
->name
, ctxt
->input
->cur
+2,
3109 xmlStrlen(ctxt
->name
)) == 0)
3113 htmlParseErr(ctxt
, XML_ERR_TAG_NAME_MISMATCH
,
3114 "Element %s embeds close tag\n",
3118 if (((NXT(2) >= 'A') && (NXT(2) <= 'Z')) ||
3119 ((NXT(2) >= 'a') && (NXT(2) <= 'z')))
3126 COPY_BUF(l
,buf
,nbchar
,cur
);
3128 htmlParseErrInt(ctxt
, XML_ERR_INVALID_CHAR
,
3129 "Invalid char in CDATA 0x%X\n", cur
);
3131 if (nbchar
>= HTML_PARSER_BIG_BUFFER_SIZE
) {
3133 if (ctxt
->sax
->cdataBlock
!= NULL
) {
3135 * Insert as CDATA, which is the same as HTML_PRESERVE_NODE
3137 ctxt
->sax
->cdataBlock(ctxt
->userData
, buf
, nbchar
);
3138 } else if (ctxt
->sax
->characters
!= NULL
) {
3139 ctxt
->sax
->characters(ctxt
->userData
, buf
, nbchar
);
3148 if ((nbchar
!= 0) && (ctxt
->sax
!= NULL
) && (!ctxt
->disableSAX
)) {
3150 if (ctxt
->sax
->cdataBlock
!= NULL
) {
3152 * Insert as CDATA, which is the same as HTML_PRESERVE_NODE
3154 ctxt
->sax
->cdataBlock(ctxt
->userData
, buf
, nbchar
);
3155 } else if (ctxt
->sax
->characters
!= NULL
) {
3156 ctxt
->sax
->characters(ctxt
->userData
, buf
, nbchar
);
3163 * htmlParseCharDataInternal:
3164 * @ctxt: an HTML parser context
3165 * @readahead: optional read ahead character in ascii range
3167 * parse a CharData section.
3168 * if we are within a CDATA section ']]>' marks an end of section.
3170 * [14] CharData ::= [^<&]* - ([^<&]* ']]>' [^<&]*)
3174 htmlParseCharDataInternal(htmlParserCtxtPtr ctxt
, int readahead
) {
3175 xmlChar buf
[HTML_PARSER_BIG_BUFFER_SIZE
+ 6];
3181 buf
[nbchar
++] = readahead
;
3185 while (((cur
!= '<') || (ctxt
->token
== '<')) &&
3186 ((cur
!= '&') || (ctxt
->token
== '&')) &&
3188 if (!(IS_CHAR(cur
))) {
3189 htmlParseErrInt(ctxt
, XML_ERR_INVALID_CHAR
,
3190 "Invalid char in CDATA 0x%X\n", cur
);
3192 COPY_BUF(l
,buf
,nbchar
,cur
);
3194 if (nbchar
>= HTML_PARSER_BIG_BUFFER_SIZE
) {
3198 * Ok the segment is to be consumed as chars.
3200 if ((ctxt
->sax
!= NULL
) && (!ctxt
->disableSAX
)) {
3201 if (areBlanks(ctxt
, buf
, nbchar
)) {
3202 if (ctxt
->keepBlanks
) {
3203 if (ctxt
->sax
->characters
!= NULL
)
3204 ctxt
->sax
->characters(ctxt
->userData
, buf
, nbchar
);
3206 if (ctxt
->sax
->ignorableWhitespace
!= NULL
)
3207 ctxt
->sax
->ignorableWhitespace(ctxt
->userData
,
3211 htmlCheckParagraph(ctxt
);
3212 if (ctxt
->sax
->characters
!= NULL
)
3213 ctxt
->sax
->characters(ctxt
->userData
, buf
, nbchar
);
3220 if (chunk
> HTML_PARSER_BUFFER_SIZE
) {
3236 * Ok the segment is to be consumed as chars.
3238 if ((ctxt
->sax
!= NULL
) && (!ctxt
->disableSAX
)) {
3239 if (areBlanks(ctxt
, buf
, nbchar
)) {
3240 if (ctxt
->keepBlanks
) {
3241 if (ctxt
->sax
->characters
!= NULL
)
3242 ctxt
->sax
->characters(ctxt
->userData
, buf
, nbchar
);
3244 if (ctxt
->sax
->ignorableWhitespace
!= NULL
)
3245 ctxt
->sax
->ignorableWhitespace(ctxt
->userData
,
3249 htmlCheckParagraph(ctxt
);
3250 if (ctxt
->sax
->characters
!= NULL
)
3251 ctxt
->sax
->characters(ctxt
->userData
, buf
, nbchar
);
3259 ctxt
->instate
= XML_PARSER_EOF
;
3264 * htmlParseCharData:
3265 * @ctxt: an HTML parser context
3267 * parse a CharData section.
3268 * if we are within a CDATA section ']]>' marks an end of section.
3270 * [14] CharData ::= [^<&]* - ([^<&]* ']]>' [^<&]*)
3274 htmlParseCharData(htmlParserCtxtPtr ctxt
) {
3275 htmlParseCharDataInternal(ctxt
, 0);
3279 * htmlParseExternalID:
3280 * @ctxt: an HTML parser context
3281 * @publicID: a xmlChar** receiving PubidLiteral
3283 * Parse an External ID or a Public ID
3285 * [75] ExternalID ::= 'SYSTEM' S SystemLiteral
3286 * | 'PUBLIC' S PubidLiteral S SystemLiteral
3288 * [83] PublicID ::= 'PUBLIC' S PubidLiteral
3290 * Returns the function returns SystemLiteral and in the second
3291 * case publicID receives PubidLiteral, is strict is off
3292 * it is possible to return NULL and have publicID set.
3296 htmlParseExternalID(htmlParserCtxtPtr ctxt
, xmlChar
**publicID
) {
3297 xmlChar
*URI
= NULL
;
3299 if ((UPPER
== 'S') && (UPP(1) == 'Y') &&
3300 (UPP(2) == 'S') && (UPP(3) == 'T') &&
3301 (UPP(4) == 'E') && (UPP(5) == 'M')) {
3303 if (!IS_BLANK_CH(CUR
)) {
3304 htmlParseErr(ctxt
, XML_ERR_SPACE_REQUIRED
,
3305 "Space required after 'SYSTEM'\n", NULL
, NULL
);
3308 URI
= htmlParseSystemLiteral(ctxt
);
3310 htmlParseErr(ctxt
, XML_ERR_URI_REQUIRED
,
3311 "htmlParseExternalID: SYSTEM, no URI\n", NULL
, NULL
);
3313 } else if ((UPPER
== 'P') && (UPP(1) == 'U') &&
3314 (UPP(2) == 'B') && (UPP(3) == 'L') &&
3315 (UPP(4) == 'I') && (UPP(5) == 'C')) {
3317 if (!IS_BLANK_CH(CUR
)) {
3318 htmlParseErr(ctxt
, XML_ERR_SPACE_REQUIRED
,
3319 "Space required after 'PUBLIC'\n", NULL
, NULL
);
3322 *publicID
= htmlParsePubidLiteral(ctxt
);
3323 if (*publicID
== NULL
) {
3324 htmlParseErr(ctxt
, XML_ERR_PUBID_REQUIRED
,
3325 "htmlParseExternalID: PUBLIC, no Public Identifier\n",
3329 if ((CUR
== '"') || (CUR
== '\'')) {
3330 URI
= htmlParseSystemLiteral(ctxt
);
3338 * @ctxt: an XML parser context
3340 * parse an XML Processing Instruction.
3342 * [16] PI ::= '<?' PITarget (S (Char* - (Char* '?>' Char*)))? '?>'
3345 htmlParsePI(htmlParserCtxtPtr ctxt
) {
3346 xmlChar
*buf
= NULL
;
3348 int size
= HTML_PARSER_BUFFER_SIZE
;
3350 const xmlChar
*target
;
3351 xmlParserInputState state
;
3354 if ((RAW
== '<') && (NXT(1) == '?')) {
3355 state
= ctxt
->instate
;
3356 ctxt
->instate
= XML_PARSER_PI
;
3358 * this is a Processing Instruction.
3364 * Parse the target name and check for special support like
3367 target
= htmlParseName(ctxt
);
3368 if (target
!= NULL
) {
3375 if ((ctxt
->sax
) && (!ctxt
->disableSAX
) &&
3376 (ctxt
->sax
->processingInstruction
!= NULL
))
3377 ctxt
->sax
->processingInstruction(ctxt
->userData
,
3379 ctxt
->instate
= state
;
3382 buf
= (xmlChar
*) xmlMallocAtomic(size
* sizeof(xmlChar
));
3384 htmlErrMemory(ctxt
, NULL
);
3385 ctxt
->instate
= state
;
3389 if (!IS_BLANK(cur
)) {
3390 htmlParseErr(ctxt
, XML_ERR_SPACE_REQUIRED
,
3391 "ParsePI: PI %s space expected\n", target
, NULL
);
3395 while ((cur
!= 0) && (cur
!= '>')) {
3396 if (len
+ 5 >= size
) {
3400 tmp
= (xmlChar
*) xmlRealloc(buf
, size
* sizeof(xmlChar
));
3402 htmlErrMemory(ctxt
, NULL
);
3404 ctxt
->instate
= state
;
3415 COPY_BUF(l
,buf
,len
,cur
);
3417 htmlParseErrInt(ctxt
, XML_ERR_INVALID_CHAR
,
3418 "Invalid char in processing instruction "
3431 htmlParseErr(ctxt
, XML_ERR_PI_NOT_FINISHED
,
3432 "ParsePI: PI %s never end ...\n", target
, NULL
);
3439 if ((ctxt
->sax
) && (!ctxt
->disableSAX
) &&
3440 (ctxt
->sax
->processingInstruction
!= NULL
))
3441 ctxt
->sax
->processingInstruction(ctxt
->userData
,
3446 htmlParseErr(ctxt
, XML_ERR_PI_NOT_STARTED
,
3447 "PI is not started correctly", NULL
, NULL
);
3449 ctxt
->instate
= state
;
3455 * @ctxt: an HTML parser context
3457 * Parse an XML (SGML) comment <!-- .... -->
3459 * [15] Comment ::= '<!--' ((Char - '-') | ('-' (Char - '-')))* '-->'
3462 htmlParseComment(htmlParserCtxtPtr ctxt
) {
3463 xmlChar
*buf
= NULL
;
3465 int size
= HTML_PARSER_BUFFER_SIZE
;
3470 xmlParserInputState state
;
3473 * Check that there is a comment right here.
3475 if ((RAW
!= '<') || (NXT(1) != '!') ||
3476 (NXT(2) != '-') || (NXT(3) != '-')) return;
3478 state
= ctxt
->instate
;
3479 ctxt
->instate
= XML_PARSER_COMMENT
;
3482 buf
= (xmlChar
*) xmlMallocAtomic(size
* sizeof(xmlChar
));
3484 htmlErrMemory(ctxt
, "buffer allocation failed\n");
3485 ctxt
->instate
= state
;
3494 htmlParseErr(ctxt
, XML_ERR_COMMENT_ABRUPTLY_ENDED
, "Comment abruptly ended", NULL
, NULL
);
3502 if (q
== '-' && r
== '>') {
3503 htmlParseErr(ctxt
, XML_ERR_COMMENT_ABRUPTLY_ENDED
, "Comment abruptly ended", NULL
, NULL
);
3509 while ((cur
!= 0) &&
3511 (r
!= '-') || (q
!= '-'))) {
3513 next
= CUR_CHAR(nl
);
3517 next
= CUR_CHAR(nl
);
3520 if ((q
== '-') && (r
== '-') && (cur
== '!') && (next
== '>')) {
3521 htmlParseErr(ctxt
, XML_ERR_COMMENT_NOT_FINISHED
,
3522 "Comment incorrectly closed by '--!>'", NULL
, NULL
);
3527 if (len
+ 5 >= size
) {
3531 tmp
= (xmlChar
*) xmlRealloc(buf
, size
* sizeof(xmlChar
));
3534 htmlErrMemory(ctxt
, "growing buffer failed\n");
3535 ctxt
->instate
= state
;
3541 COPY_BUF(ql
,buf
,len
,q
);
3543 htmlParseErrInt(ctxt
, XML_ERR_INVALID_CHAR
,
3544 "Invalid char in comment 0x%X\n", q
);
3558 if ((ctxt
->sax
!= NULL
) && (ctxt
->sax
->comment
!= NULL
) &&
3559 (!ctxt
->disableSAX
))
3560 ctxt
->sax
->comment(ctxt
->userData
, buf
);
3562 ctxt
->instate
= state
;
3567 htmlParseErr(ctxt
, XML_ERR_COMMENT_NOT_FINISHED
,
3568 "Comment not terminated \n<!--%.50s\n", buf
, NULL
);
3574 * @ctxt: an HTML parser context
3576 * parse Reference declarations
3578 * [66] CharRef ::= '&#' [0-9]+ ';' |
3579 * '&#x' [0-9a-fA-F]+ ';'
3581 * Returns the value parsed (as an int)
3584 htmlParseCharRef(htmlParserCtxtPtr ctxt
) {
3587 if ((ctxt
== NULL
) || (ctxt
->input
== NULL
)) {
3588 htmlParseErr(ctxt
, XML_ERR_INTERNAL_ERROR
,
3589 "htmlParseCharRef: context error\n",
3593 if ((CUR
== '&') && (NXT(1) == '#') &&
3594 ((NXT(2) == 'x') || NXT(2) == 'X')) {
3596 while (CUR
!= ';') {
3597 if ((CUR
>= '0') && (CUR
<= '9')) {
3599 val
= val
* 16 + (CUR
- '0');
3600 } else if ((CUR
>= 'a') && (CUR
<= 'f')) {
3602 val
= val
* 16 + (CUR
- 'a') + 10;
3603 } else if ((CUR
>= 'A') && (CUR
<= 'F')) {
3605 val
= val
* 16 + (CUR
- 'A') + 10;
3607 htmlParseErr(ctxt
, XML_ERR_INVALID_HEX_CHARREF
,
3608 "htmlParseCharRef: missing semicolon\n",
3616 } else if ((CUR
== '&') && (NXT(1) == '#')) {
3618 while (CUR
!= ';') {
3619 if ((CUR
>= '0') && (CUR
<= '9')) {
3621 val
= val
* 10 + (CUR
- '0');
3623 htmlParseErr(ctxt
, XML_ERR_INVALID_DEC_CHARREF
,
3624 "htmlParseCharRef: missing semicolon\n",
3633 htmlParseErr(ctxt
, XML_ERR_INVALID_CHARREF
,
3634 "htmlParseCharRef: invalid value\n", NULL
, NULL
);
3637 * Check the value IS_CHAR ...
3641 } else if (val
>= 0x110000) {
3642 htmlParseErr(ctxt
, XML_ERR_INVALID_CHAR
,
3643 "htmlParseCharRef: value too large\n", NULL
, NULL
);
3645 htmlParseErrInt(ctxt
, XML_ERR_INVALID_CHAR
,
3646 "htmlParseCharRef: invalid xmlChar value %d\n",
3654 * htmlParseDocTypeDecl:
3655 * @ctxt: an HTML parser context
3657 * parse a DOCTYPE declaration
3659 * [28] doctypedecl ::= '<!DOCTYPE' S Name (S ExternalID)? S?
3660 * ('[' (markupdecl | PEReference | S)* ']' S?)? '>'
3664 htmlParseDocTypeDecl(htmlParserCtxtPtr ctxt
) {
3665 const xmlChar
*name
;
3666 xmlChar
*ExternalID
= NULL
;
3667 xmlChar
*URI
= NULL
;
3670 * We know that '<!DOCTYPE' has been detected.
3677 * Parse the DOCTYPE name.
3679 name
= htmlParseName(ctxt
);
3681 htmlParseErr(ctxt
, XML_ERR_NAME_REQUIRED
,
3682 "htmlParseDocTypeDecl : no DOCTYPE name !\n",
3686 * Check that upper(name) == "HTML" !!!!!!!!!!!!!
3692 * Check for SystemID and ExternalID
3694 URI
= htmlParseExternalID(ctxt
, &ExternalID
);
3698 * We should be at the end of the DOCTYPE declaration.
3701 htmlParseErr(ctxt
, XML_ERR_DOCTYPE_NOT_FINISHED
,
3702 "DOCTYPE improperly terminated\n", NULL
, NULL
);
3703 /* Ignore bogus content */
3704 while ((CUR
!= 0) && (CUR
!= '>'))
3711 * Create or update the document accordingly to the DOCTYPE
3713 if ((ctxt
->sax
!= NULL
) && (ctxt
->sax
->internalSubset
!= NULL
) &&
3714 (!ctxt
->disableSAX
))
3715 ctxt
->sax
->internalSubset(ctxt
->userData
, name
, ExternalID
, URI
);
3718 * Cleanup, since we don't use all those identifiers
3720 if (URI
!= NULL
) xmlFree(URI
);
3721 if (ExternalID
!= NULL
) xmlFree(ExternalID
);
3725 * htmlParseAttribute:
3726 * @ctxt: an HTML parser context
3727 * @value: a xmlChar ** used to store the value of the attribute
3729 * parse an attribute
3731 * [41] Attribute ::= Name Eq AttValue
3733 * [25] Eq ::= S? '=' S?
3737 * [NS 11] Attribute ::= QName Eq AttValue
3739 * Also the case QName == xmlns:??? is handled independently as a namespace
3742 * Returns the attribute name, and the value in *value.
3745 static const xmlChar
*
3746 htmlParseAttribute(htmlParserCtxtPtr ctxt
, xmlChar
**value
) {
3747 const xmlChar
*name
;
3748 xmlChar
*val
= NULL
;
3751 name
= htmlParseHTMLName(ctxt
);
3753 htmlParseErr(ctxt
, XML_ERR_NAME_REQUIRED
,
3754 "error parsing attribute name\n", NULL
, NULL
);
3765 val
= htmlParseAttValue(ctxt
);
3773 * htmlCheckEncodingDirect:
3774 * @ctxt: an HTML parser context
3775 * @attvalue: the attribute value
3777 * Checks an attribute value to detect
3779 * If a new encoding is detected the parser is switched to decode
3783 htmlCheckEncodingDirect(htmlParserCtxtPtr ctxt
, const xmlChar
*encoding
) {
3785 if ((ctxt
== NULL
) || (encoding
== NULL
) ||
3786 (ctxt
->options
& HTML_PARSE_IGNORE_ENC
))
3789 /* do not change encoding */
3790 if (ctxt
->input
->encoding
!= NULL
)
3793 if (encoding
!= NULL
) {
3794 xmlCharEncoding enc
;
3795 xmlCharEncodingHandlerPtr handler
;
3797 while ((*encoding
== ' ') || (*encoding
== '\t')) encoding
++;
3799 if (ctxt
->input
->encoding
!= NULL
)
3800 xmlFree((xmlChar
*) ctxt
->input
->encoding
);
3801 ctxt
->input
->encoding
= xmlStrdup(encoding
);
3803 enc
= xmlParseCharEncoding((const char *) encoding
);
3805 * registered set of known encodings
3807 if (enc
!= XML_CHAR_ENCODING_ERROR
) {
3808 if (((enc
== XML_CHAR_ENCODING_UTF16LE
) ||
3809 (enc
== XML_CHAR_ENCODING_UTF16BE
) ||
3810 (enc
== XML_CHAR_ENCODING_UCS4LE
) ||
3811 (enc
== XML_CHAR_ENCODING_UCS4BE
)) &&
3812 (ctxt
->input
->buf
!= NULL
) &&
3813 (ctxt
->input
->buf
->encoder
== NULL
)) {
3814 htmlParseErr(ctxt
, XML_ERR_INVALID_ENCODING
,
3815 "htmlCheckEncoding: wrong encoding meta\n",
3818 xmlSwitchEncoding(ctxt
, enc
);
3820 ctxt
->charset
= XML_CHAR_ENCODING_UTF8
;
3823 * fallback for unknown encodings
3825 handler
= xmlFindCharEncodingHandler((const char *) encoding
);
3826 if (handler
!= NULL
) {
3827 xmlSwitchToEncoding(ctxt
, handler
);
3828 ctxt
->charset
= XML_CHAR_ENCODING_UTF8
;
3830 htmlParseErr(ctxt
, XML_ERR_UNSUPPORTED_ENCODING
,
3831 "htmlCheckEncoding: unknown encoding %s\n",
3836 if ((ctxt
->input
->buf
!= NULL
) &&
3837 (ctxt
->input
->buf
->encoder
!= NULL
) &&
3838 (ctxt
->input
->buf
->raw
!= NULL
) &&
3839 (ctxt
->input
->buf
->buffer
!= NULL
)) {
3844 * convert as much as possible to the parser reading buffer.
3846 processed
= ctxt
->input
->cur
- ctxt
->input
->base
;
3847 xmlBufShrink(ctxt
->input
->buf
->buffer
, processed
);
3848 nbchars
= xmlCharEncInput(ctxt
->input
->buf
, 1);
3849 xmlBufResetInput(ctxt
->input
->buf
->buffer
, ctxt
->input
);
3851 htmlParseErr(ctxt
, XML_ERR_INVALID_ENCODING
,
3852 "htmlCheckEncoding: encoder error\n",
3860 * htmlCheckEncoding:
3861 * @ctxt: an HTML parser context
3862 * @attvalue: the attribute value
3864 * Checks an http-equiv attribute from a Meta tag to detect
3866 * If a new encoding is detected the parser is switched to decode
3870 htmlCheckEncoding(htmlParserCtxtPtr ctxt
, const xmlChar
*attvalue
) {
3871 const xmlChar
*encoding
;
3876 encoding
= xmlStrcasestr(attvalue
, BAD_CAST
"charset");
3877 if (encoding
!= NULL
) {
3883 if (encoding
&& IS_BLANK_CH(*encoding
))
3884 encoding
= xmlStrcasestr(attvalue
, BAD_CAST
"=");
3885 if (encoding
&& *encoding
== '=') {
3887 htmlCheckEncodingDirect(ctxt
, encoding
);
3893 * @ctxt: an HTML parser context
3894 * @atts: the attributes values
3896 * Checks an attributes from a Meta tag
3899 htmlCheckMeta(htmlParserCtxtPtr ctxt
, const xmlChar
**atts
) {
3901 const xmlChar
*att
, *value
;
3903 const xmlChar
*content
= NULL
;
3905 if ((ctxt
== NULL
) || (atts
== NULL
))
3910 while (att
!= NULL
) {
3912 if ((value
!= NULL
) && (!xmlStrcasecmp(att
, BAD_CAST
"http-equiv"))
3913 && (!xmlStrcasecmp(value
, BAD_CAST
"Content-Type")))
3915 else if ((value
!= NULL
) && (!xmlStrcasecmp(att
, BAD_CAST
"charset")))
3916 htmlCheckEncodingDirect(ctxt
, value
);
3917 else if ((value
!= NULL
) && (!xmlStrcasecmp(att
, BAD_CAST
"content")))
3921 if ((http
) && (content
!= NULL
))
3922 htmlCheckEncoding(ctxt
, content
);
3927 * htmlParseStartTag:
3928 * @ctxt: an HTML parser context
3930 * parse a start of tag either for rule element or
3931 * EmptyElement. In both case we don't parse the tag closing chars.
3933 * [40] STag ::= '<' Name (S Attribute)* S? '>'
3935 * [44] EmptyElemTag ::= '<' Name (S Attribute)* S? '/>'
3939 * [NS 8] STag ::= '<' QName (S Attribute)* S? '>'
3941 * [NS 10] EmptyElement ::= '<' QName (S Attribute)* S? '/>'
3943 * Returns 0 in case of success, -1 in case of error and 1 if discarded
3947 htmlParseStartTag(htmlParserCtxtPtr ctxt
) {
3948 const xmlChar
*name
;
3949 const xmlChar
*attname
;
3951 const xmlChar
**atts
;
3958 if ((ctxt
== NULL
) || (ctxt
->input
== NULL
)) {
3959 htmlParseErr(ctxt
, XML_ERR_INTERNAL_ERROR
,
3960 "htmlParseStartTag: context error\n", NULL
, NULL
);
3963 if (ctxt
->instate
== XML_PARSER_EOF
)
3965 if (CUR
!= '<') return -1;
3969 maxatts
= ctxt
->maxatts
;
3972 name
= htmlParseHTMLName(ctxt
);
3974 htmlParseErr(ctxt
, XML_ERR_NAME_REQUIRED
,
3975 "htmlParseStartTag: invalid element name\n",
3977 /* Dump the bogus tag like browsers do */
3978 while ((CUR
!= 0) && (CUR
!= '>') &&
3979 (ctxt
->instate
!= XML_PARSER_EOF
))
3983 if (xmlStrEqual(name
, BAD_CAST
"meta"))
3987 * Check for auto-closure of HTML elements.
3989 htmlAutoClose(ctxt
, name
);
3992 * Check for implied HTML elements.
3994 htmlCheckImplied(ctxt
, name
);
3997 * Avoid html at any level > 0, head at any level != 1
3998 * or any attempt to recurse body
4000 if ((ctxt
->nameNr
> 0) && (xmlStrEqual(name
, BAD_CAST
"html"))) {
4001 htmlParseErr(ctxt
, XML_HTML_STRUCURE_ERROR
,
4002 "htmlParseStartTag: misplaced <html> tag\n",
4007 if ((ctxt
->nameNr
!= 1) &&
4008 (xmlStrEqual(name
, BAD_CAST
"head"))) {
4009 htmlParseErr(ctxt
, XML_HTML_STRUCURE_ERROR
,
4010 "htmlParseStartTag: misplaced <head> tag\n",
4015 if (xmlStrEqual(name
, BAD_CAST
"body")) {
4017 for (indx
= 0;indx
< ctxt
->nameNr
;indx
++) {
4018 if (xmlStrEqual(ctxt
->nameTab
[indx
], BAD_CAST
"body")) {
4019 htmlParseErr(ctxt
, XML_HTML_STRUCURE_ERROR
,
4020 "htmlParseStartTag: misplaced <body> tag\n",
4029 * Now parse the attributes, it ends up with the ending
4034 while ((CUR
!= 0) &&
4036 ((CUR
!= '/') || (NXT(1) != '>'))) {
4038 attname
= htmlParseAttribute(ctxt
, &attvalue
);
4039 if (attname
!= NULL
) {
4042 * Well formedness requires at most one declaration of an attribute
4044 for (i
= 0; i
< nbatts
;i
+= 2) {
4045 if (xmlStrEqual(atts
[i
], attname
)) {
4046 htmlParseErr(ctxt
, XML_ERR_ATTRIBUTE_REDEFINED
,
4047 "Attribute %s redefined\n", attname
, NULL
);
4048 if (attvalue
!= NULL
)
4055 * Add the pair to atts
4058 maxatts
= 22; /* allow for 10 attrs by default */
4059 atts
= (const xmlChar
**)
4060 xmlMalloc(maxatts
* sizeof(xmlChar
*));
4062 htmlErrMemory(ctxt
, NULL
);
4063 if (attvalue
!= NULL
)
4068 ctxt
->maxatts
= maxatts
;
4069 } else if (nbatts
+ 4 > maxatts
) {
4073 n
= (const xmlChar
**) xmlRealloc((void *) atts
,
4074 maxatts
* sizeof(const xmlChar
*));
4076 htmlErrMemory(ctxt
, NULL
);
4077 if (attvalue
!= NULL
)
4083 ctxt
->maxatts
= maxatts
;
4085 atts
[nbatts
++] = attname
;
4086 atts
[nbatts
++] = attvalue
;
4087 atts
[nbatts
] = NULL
;
4088 atts
[nbatts
+ 1] = NULL
;
4091 if (attvalue
!= NULL
)
4093 /* Dump the bogus attribute string up to the next blank or
4094 * the end of the tag. */
4095 while ((CUR
!= 0) &&
4096 !(IS_BLANK_CH(CUR
)) && (CUR
!= '>') &&
4097 ((CUR
!= '/') || (NXT(1) != '>')))
4106 * Handle specific association to the META tag
4108 if (meta
&& (nbatts
!= 0))
4109 htmlCheckMeta(ctxt
, atts
);
4112 * SAX: Start of Element !
4115 htmlnamePush(ctxt
, name
);
4116 if ((ctxt
->sax
!= NULL
) && (ctxt
->sax
->startElement
!= NULL
)) {
4118 ctxt
->sax
->startElement(ctxt
->userData
, name
, atts
);
4120 ctxt
->sax
->startElement(ctxt
->userData
, name
, NULL
);
4125 for (i
= 1;i
< nbatts
;i
+= 2) {
4126 if (atts
[i
] != NULL
)
4127 xmlFree((xmlChar
*) atts
[i
]);
4136 * @ctxt: an HTML parser context
4138 * parse an end of tag
4140 * [42] ETag ::= '</' Name S? '>'
4144 * [NS 9] ETag ::= '</' QName S? '>'
4146 * Returns 1 if the current level should be closed.
4150 htmlParseEndTag(htmlParserCtxtPtr ctxt
)
4152 const xmlChar
*name
;
4153 const xmlChar
*oldname
;
4156 if ((CUR
!= '<') || (NXT(1) != '/')) {
4157 htmlParseErr(ctxt
, XML_ERR_LTSLASH_REQUIRED
,
4158 "htmlParseEndTag: '</' not found\n", NULL
, NULL
);
4163 name
= htmlParseHTMLName(ctxt
);
4167 * We should definitely be at the ending "S? '>'" part
4171 htmlParseErr(ctxt
, XML_ERR_GT_REQUIRED
,
4172 "End tag : expected '>'\n", NULL
, NULL
);
4173 /* Skip to next '>' */
4174 while ((CUR
!= 0) && (CUR
!= '>'))
4181 * if we ignored misplaced tags in htmlParseStartTag don't pop them
4184 if ((ctxt
->depth
> 0) &&
4185 (xmlStrEqual(name
, BAD_CAST
"html") ||
4186 xmlStrEqual(name
, BAD_CAST
"body") ||
4187 xmlStrEqual(name
, BAD_CAST
"head"))) {
4193 * If the name read is not one of the element in the parsing stack
4194 * then return, it's just an error.
4196 for (i
= (ctxt
->nameNr
- 1); i
>= 0; i
--) {
4197 if (xmlStrEqual(name
, ctxt
->nameTab
[i
]))
4201 htmlParseErr(ctxt
, XML_ERR_TAG_NAME_MISMATCH
,
4202 "Unexpected end tag : %s\n", name
, NULL
);
4208 * Check for auto-closure of HTML elements.
4211 htmlAutoCloseOnClose(ctxt
, name
);
4214 * Well formedness constraints, opening and closing must match.
4215 * With the exception that the autoclose may have popped stuff out
4218 if ((ctxt
->name
!= NULL
) && (!xmlStrEqual(ctxt
->name
, name
))) {
4219 htmlParseErr(ctxt
, XML_ERR_TAG_NAME_MISMATCH
,
4220 "Opening and ending tag mismatch: %s and %s\n",
4227 oldname
= ctxt
->name
;
4228 if ((oldname
!= NULL
) && (xmlStrEqual(oldname
, name
))) {
4229 if ((ctxt
->sax
!= NULL
) && (ctxt
->sax
->endElement
!= NULL
))
4230 ctxt
->sax
->endElement(ctxt
->userData
, name
);
4231 htmlNodeInfoPop(ctxt
);
4243 * htmlParseReference:
4244 * @ctxt: an HTML parser context
4246 * parse and handle entity references in content,
4247 * this will end-up in a call to character() since this is either a
4248 * CharRef, or a predefined entity.
4251 htmlParseReference(htmlParserCtxtPtr ctxt
) {
4252 const htmlEntityDesc
* ent
;
4254 const xmlChar
*name
;
4255 if (CUR
!= '&') return;
4257 if (NXT(1) == '#') {
4261 c
= htmlParseCharRef(ctxt
);
4265 if (c
< 0x80) { out
[i
++]= c
; bits
= -6; }
4266 else if (c
< 0x800) { out
[i
++]=((c
>> 6) & 0x1F) | 0xC0; bits
= 0; }
4267 else if (c
< 0x10000) { out
[i
++]=((c
>> 12) & 0x0F) | 0xE0; bits
= 6; }
4268 else { out
[i
++]=((c
>> 18) & 0x07) | 0xF0; bits
= 12; }
4270 for ( ; bits
>= 0; bits
-= 6) {
4271 out
[i
++]= ((c
>> bits
) & 0x3F) | 0x80;
4275 htmlCheckParagraph(ctxt
);
4276 if ((ctxt
->sax
!= NULL
) && (ctxt
->sax
->characters
!= NULL
))
4277 ctxt
->sax
->characters(ctxt
->userData
, out
, i
);
4279 ent
= htmlParseEntityRef(ctxt
, &name
);
4281 htmlCheckParagraph(ctxt
);
4282 if ((ctxt
->sax
!= NULL
) && (ctxt
->sax
->characters
!= NULL
))
4283 ctxt
->sax
->characters(ctxt
->userData
, BAD_CAST
"&", 1);
4286 if ((ent
== NULL
) || !(ent
->value
> 0)) {
4287 htmlCheckParagraph(ctxt
);
4288 if ((ctxt
->sax
!= NULL
) && (ctxt
->sax
->characters
!= NULL
)) {
4289 ctxt
->sax
->characters(ctxt
->userData
, BAD_CAST
"&", 1);
4290 ctxt
->sax
->characters(ctxt
->userData
, name
, xmlStrlen(name
));
4291 /* ctxt->sax->characters(ctxt->userData, BAD_CAST ";", 1); */
4299 { out
[i
++]= c
; bits
= -6; }
4301 { out
[i
++]=((c
>> 6) & 0x1F) | 0xC0; bits
= 0; }
4302 else if (c
< 0x10000)
4303 { out
[i
++]=((c
>> 12) & 0x0F) | 0xE0; bits
= 6; }
4305 { out
[i
++]=((c
>> 18) & 0x07) | 0xF0; bits
= 12; }
4307 for ( ; bits
>= 0; bits
-= 6) {
4308 out
[i
++]= ((c
>> bits
) & 0x3F) | 0x80;
4312 htmlCheckParagraph(ctxt
);
4313 if ((ctxt
->sax
!= NULL
) && (ctxt
->sax
->characters
!= NULL
))
4314 ctxt
->sax
->characters(ctxt
->userData
, out
, i
);
4321 * @ctxt: an HTML parser context
4323 * Parse a content: comment, sub-element, reference or text.
4324 * Kept for compatibility with old code
4328 htmlParseContent(htmlParserCtxtPtr ctxt
) {
4329 xmlChar
*currentNode
;
4331 const xmlChar
*name
;
4333 currentNode
= xmlStrdup(ctxt
->name
);
4334 depth
= ctxt
->nameNr
;
4338 if (ctxt
->instate
== XML_PARSER_EOF
)
4342 * Our tag or one of it's parent or children is ending.
4344 if ((CUR
== '<') && (NXT(1) == '/')) {
4345 if (htmlParseEndTag(ctxt
) &&
4346 ((currentNode
!= NULL
) || (ctxt
->nameNr
== 0))) {
4347 if (currentNode
!= NULL
)
4348 xmlFree(currentNode
);
4351 continue; /* while */
4354 else if ((CUR
== '<') &&
4355 ((IS_ASCII_LETTER(NXT(1))) ||
4356 (NXT(1) == '_') || (NXT(1) == ':'))) {
4357 name
= htmlParseHTMLName_nonInvasive(ctxt
);
4359 htmlParseErr(ctxt
, XML_ERR_NAME_REQUIRED
,
4360 "htmlParseStartTag: invalid element name\n",
4362 /* Dump the bogus tag like browsers do */
4363 while ((CUR
!= 0) && (CUR
!= '>'))
4366 if (currentNode
!= NULL
)
4367 xmlFree(currentNode
);
4371 if (ctxt
->name
!= NULL
) {
4372 if (htmlCheckAutoClose(name
, ctxt
->name
) == 1) {
4373 htmlAutoClose(ctxt
, name
);
4380 * Has this node been popped out during parsing of
4383 if ((ctxt
->nameNr
> 0) && (depth
>= ctxt
->nameNr
) &&
4384 (!xmlStrEqual(currentNode
, ctxt
->name
)))
4386 if (currentNode
!= NULL
) xmlFree(currentNode
);
4390 if ((CUR
!= 0) && ((xmlStrEqual(currentNode
, BAD_CAST
"script")) ||
4391 (xmlStrEqual(currentNode
, BAD_CAST
"style")))) {
4393 * Handle SCRIPT/STYLE separately
4395 htmlParseScript(ctxt
);
4398 else if ((CUR
== '<') && (NXT(1) == '!')) {
4400 * Sometimes DOCTYPE arrives in the middle of the document
4402 if ((UPP(2) == 'D') && (UPP(3) == 'O') &&
4403 (UPP(4) == 'C') && (UPP(5) == 'T') &&
4404 (UPP(6) == 'Y') && (UPP(7) == 'P') &&
4406 htmlParseErr(ctxt
, XML_HTML_STRUCURE_ERROR
,
4407 "Misplaced DOCTYPE declaration\n",
4408 BAD_CAST
"DOCTYPE" , NULL
);
4409 htmlParseDocTypeDecl(ctxt
);
4412 * First case : a comment
4414 else if ((NXT(2) == '-') && (NXT(3) == '-')) {
4415 htmlParseComment(ctxt
);
4418 htmlSkipBogusComment(ctxt
);
4423 * Second case : a Processing Instruction.
4425 else if ((CUR
== '<') && (NXT(1) == '?')) {
4430 * Third case : a sub-element.
4432 else if ((CUR
== '<') && IS_ASCII_LETTER(NXT(1))) {
4433 htmlParseElement(ctxt
);
4435 else if (CUR
== '<') {
4436 if ((ctxt
->sax
!= NULL
) && (!ctxt
->disableSAX
) &&
4437 (ctxt
->sax
->characters
!= NULL
))
4438 ctxt
->sax
->characters(ctxt
->userData
, BAD_CAST
"<", 1);
4443 * Fourth case : a reference. If if has not been resolved,
4444 * parsing returns it's Name, create the node
4446 else if (CUR
== '&') {
4447 htmlParseReference(ctxt
);
4451 * Fifth case : end of the resource
4453 else if (CUR
== 0) {
4454 htmlAutoCloseOnEnd(ctxt
);
4459 * Last case, text. Note that References are handled directly.
4462 htmlParseCharData(ctxt
);
4466 if (currentNode
!= NULL
) xmlFree(currentNode
);
4471 * @ctxt: an HTML parser context
4473 * parse an HTML element, this is highly recursive
4474 * this is kept for compatibility with previous code versions
4476 * [39] element ::= EmptyElemTag | STag content ETag
4478 * [41] Attribute ::= Name Eq AttValue
4482 htmlParseElement(htmlParserCtxtPtr ctxt
) {
4483 const xmlChar
*name
;
4484 xmlChar
*currentNode
= NULL
;
4485 const htmlElemDesc
* info
;
4486 htmlParserNodeInfo node_info
;
4489 const xmlChar
*oldptr
;
4491 if ((ctxt
== NULL
) || (ctxt
->input
== NULL
)) {
4492 htmlParseErr(ctxt
, XML_ERR_INTERNAL_ERROR
,
4493 "htmlParseElement: context error\n", NULL
, NULL
);
4497 if (ctxt
->instate
== XML_PARSER_EOF
)
4500 /* Capture start position */
4501 if (ctxt
->record_info
) {
4502 node_info
.begin_pos
= ctxt
->input
->consumed
+
4503 (CUR_PTR
- ctxt
->input
->base
);
4504 node_info
.begin_line
= ctxt
->input
->line
;
4507 failed
= htmlParseStartTag(ctxt
);
4509 if ((failed
== -1) || (name
== NULL
)) {
4516 * Lookup the info for that element.
4518 info
= htmlTagLookup(name
);
4520 htmlParseErr(ctxt
, XML_HTML_UNKNOWN_TAG
,
4521 "Tag %s invalid\n", name
, NULL
);
4525 * Check for an Empty Element labeled the XML/SGML way
4527 if ((CUR
== '/') && (NXT(1) == '>')) {
4529 if ((ctxt
->sax
!= NULL
) && (ctxt
->sax
->endElement
!= NULL
))
4530 ctxt
->sax
->endElement(ctxt
->userData
, name
);
4538 htmlParseErr(ctxt
, XML_ERR_GT_REQUIRED
,
4539 "Couldn't find end of Start Tag %s\n", name
, NULL
);
4542 * end of parsing of this node.
4544 if (xmlStrEqual(name
, ctxt
->name
)) {
4550 * Capture end position and add node
4552 if (ctxt
->record_info
) {
4553 node_info
.end_pos
= ctxt
->input
->consumed
+
4554 (CUR_PTR
- ctxt
->input
->base
);
4555 node_info
.end_line
= ctxt
->input
->line
;
4556 node_info
.node
= ctxt
->node
;
4557 xmlParserAddNodeInfo(ctxt
, &node_info
);
4563 * Check for an Empty Element from DTD definition
4565 if ((info
!= NULL
) && (info
->empty
)) {
4566 if ((ctxt
->sax
!= NULL
) && (ctxt
->sax
->endElement
!= NULL
))
4567 ctxt
->sax
->endElement(ctxt
->userData
, name
);
4573 * Parse the content of the element:
4575 currentNode
= xmlStrdup(ctxt
->name
);
4576 depth
= ctxt
->nameNr
;
4578 oldptr
= ctxt
->input
->cur
;
4579 htmlParseContent(ctxt
);
4580 if (oldptr
==ctxt
->input
->cur
) break;
4581 if (ctxt
->nameNr
< depth
) break;
4585 * Capture end position and add node
4587 if ( currentNode
!= NULL
&& ctxt
->record_info
) {
4588 node_info
.end_pos
= ctxt
->input
->consumed
+
4589 (CUR_PTR
- ctxt
->input
->base
);
4590 node_info
.end_line
= ctxt
->input
->line
;
4591 node_info
.node
= ctxt
->node
;
4592 xmlParserAddNodeInfo(ctxt
, &node_info
);
4595 htmlAutoCloseOnEnd(ctxt
);
4598 if (currentNode
!= NULL
)
4599 xmlFree(currentNode
);
4603 htmlParserFinishElementParsing(htmlParserCtxtPtr ctxt
) {
4605 * Capture end position and add node
4607 if ( ctxt
->node
!= NULL
&& ctxt
->record_info
) {
4608 ctxt
->nodeInfo
->end_pos
= ctxt
->input
->consumed
+
4609 (CUR_PTR
- ctxt
->input
->base
);
4610 ctxt
->nodeInfo
->end_line
= ctxt
->input
->line
;
4611 ctxt
->nodeInfo
->node
= ctxt
->node
;
4612 xmlParserAddNodeInfo(ctxt
, ctxt
->nodeInfo
);
4613 htmlNodeInfoPop(ctxt
);
4616 htmlAutoCloseOnEnd(ctxt
);
4621 * htmlParseElementInternal:
4622 * @ctxt: an HTML parser context
4624 * parse an HTML element, new version, non recursive
4626 * [39] element ::= EmptyElemTag | STag content ETag
4628 * [41] Attribute ::= Name Eq AttValue
4632 htmlParseElementInternal(htmlParserCtxtPtr ctxt
) {
4633 const xmlChar
*name
;
4634 const htmlElemDesc
* info
;
4635 htmlParserNodeInfo node_info
= { NULL
, 0, 0, 0, 0 };
4638 if ((ctxt
== NULL
) || (ctxt
->input
== NULL
)) {
4639 htmlParseErr(ctxt
, XML_ERR_INTERNAL_ERROR
,
4640 "htmlParseElementInternal: context error\n", NULL
, NULL
);
4644 if (ctxt
->instate
== XML_PARSER_EOF
)
4647 /* Capture start position */
4648 if (ctxt
->record_info
) {
4649 node_info
.begin_pos
= ctxt
->input
->consumed
+
4650 (CUR_PTR
- ctxt
->input
->base
);
4651 node_info
.begin_line
= ctxt
->input
->line
;
4654 failed
= htmlParseStartTag(ctxt
);
4656 if ((failed
== -1) || (name
== NULL
)) {
4663 * Lookup the info for that element.
4665 info
= htmlTagLookup(name
);
4667 htmlParseErr(ctxt
, XML_HTML_UNKNOWN_TAG
,
4668 "Tag %s invalid\n", name
, NULL
);
4672 * Check for an Empty Element labeled the XML/SGML way
4674 if ((CUR
== '/') && (NXT(1) == '>')) {
4676 if ((ctxt
->sax
!= NULL
) && (ctxt
->sax
->endElement
!= NULL
))
4677 ctxt
->sax
->endElement(ctxt
->userData
, name
);
4685 htmlParseErr(ctxt
, XML_ERR_GT_REQUIRED
,
4686 "Couldn't find end of Start Tag %s\n", name
, NULL
);
4689 * end of parsing of this node.
4691 if (xmlStrEqual(name
, ctxt
->name
)) {
4696 if (ctxt
->record_info
)
4697 htmlNodeInfoPush(ctxt
, &node_info
);
4698 htmlParserFinishElementParsing(ctxt
);
4703 * Check for an Empty Element from DTD definition
4705 if ((info
!= NULL
) && (info
->empty
)) {
4706 if ((ctxt
->sax
!= NULL
) && (ctxt
->sax
->endElement
!= NULL
))
4707 ctxt
->sax
->endElement(ctxt
->userData
, name
);
4712 if (ctxt
->record_info
)
4713 htmlNodeInfoPush(ctxt
, &node_info
);
4717 * htmlParseContentInternal:
4718 * @ctxt: an HTML parser context
4720 * Parse a content: comment, sub-element, reference or text.
4721 * New version for non recursive htmlParseElementInternal
4725 htmlParseContentInternal(htmlParserCtxtPtr ctxt
) {
4726 xmlChar
*currentNode
;
4728 const xmlChar
*name
;
4730 currentNode
= xmlStrdup(ctxt
->name
);
4731 depth
= ctxt
->nameNr
;
4735 if (ctxt
->instate
== XML_PARSER_EOF
)
4739 * Our tag or one of it's parent or children is ending.
4741 if ((CUR
== '<') && (NXT(1) == '/')) {
4742 if (htmlParseEndTag(ctxt
) &&
4743 ((currentNode
!= NULL
) || (ctxt
->nameNr
== 0))) {
4744 if (currentNode
!= NULL
)
4745 xmlFree(currentNode
);
4747 currentNode
= xmlStrdup(ctxt
->name
);
4748 depth
= ctxt
->nameNr
;
4750 continue; /* while */
4753 else if ((CUR
== '<') &&
4754 ((IS_ASCII_LETTER(NXT(1))) ||
4755 (NXT(1) == '_') || (NXT(1) == ':'))) {
4756 name
= htmlParseHTMLName_nonInvasive(ctxt
);
4758 htmlParseErr(ctxt
, XML_ERR_NAME_REQUIRED
,
4759 "htmlParseStartTag: invalid element name\n",
4761 /* Dump the bogus tag like browsers do */
4762 while ((CUR
== 0) && (CUR
!= '>'))
4765 htmlParserFinishElementParsing(ctxt
);
4766 if (currentNode
!= NULL
)
4767 xmlFree(currentNode
);
4769 currentNode
= xmlStrdup(ctxt
->name
);
4770 depth
= ctxt
->nameNr
;
4774 if (ctxt
->name
!= NULL
) {
4775 if (htmlCheckAutoClose(name
, ctxt
->name
) == 1) {
4776 htmlAutoClose(ctxt
, name
);
4783 * Has this node been popped out during parsing of
4786 if ((ctxt
->nameNr
> 0) && (depth
>= ctxt
->nameNr
) &&
4787 (!xmlStrEqual(currentNode
, ctxt
->name
)))
4789 htmlParserFinishElementParsing(ctxt
);
4790 if (currentNode
!= NULL
) xmlFree(currentNode
);
4792 currentNode
= xmlStrdup(ctxt
->name
);
4793 depth
= ctxt
->nameNr
;
4797 if ((CUR
!= 0) && ((xmlStrEqual(currentNode
, BAD_CAST
"script")) ||
4798 (xmlStrEqual(currentNode
, BAD_CAST
"style")))) {
4800 * Handle SCRIPT/STYLE separately
4802 htmlParseScript(ctxt
);
4805 else if ((CUR
== '<') && (NXT(1) == '!')) {
4807 * Sometimes DOCTYPE arrives in the middle of the document
4809 if ((UPP(2) == 'D') && (UPP(3) == 'O') &&
4810 (UPP(4) == 'C') && (UPP(5) == 'T') &&
4811 (UPP(6) == 'Y') && (UPP(7) == 'P') &&
4813 htmlParseErr(ctxt
, XML_HTML_STRUCURE_ERROR
,
4814 "Misplaced DOCTYPE declaration\n",
4815 BAD_CAST
"DOCTYPE" , NULL
);
4816 htmlParseDocTypeDecl(ctxt
);
4819 * First case : a comment
4821 else if ((NXT(2) == '-') && (NXT(3) == '-')) {
4822 htmlParseComment(ctxt
);
4825 htmlSkipBogusComment(ctxt
);
4830 * Second case : a Processing Instruction.
4832 else if ((CUR
== '<') && (NXT(1) == '?')) {
4837 * Third case : a sub-element.
4839 else if ((CUR
== '<') && IS_ASCII_LETTER(NXT(1))) {
4840 htmlParseElementInternal(ctxt
);
4841 if (currentNode
!= NULL
) xmlFree(currentNode
);
4843 currentNode
= xmlStrdup(ctxt
->name
);
4844 depth
= ctxt
->nameNr
;
4846 else if (CUR
== '<') {
4847 if ((ctxt
->sax
!= NULL
) && (!ctxt
->disableSAX
) &&
4848 (ctxt
->sax
->characters
!= NULL
))
4849 ctxt
->sax
->characters(ctxt
->userData
, BAD_CAST
"<", 1);
4854 * Fourth case : a reference. If if has not been resolved,
4855 * parsing returns it's Name, create the node
4857 else if (CUR
== '&') {
4858 htmlParseReference(ctxt
);
4862 * Fifth case : end of the resource
4864 else if (CUR
== 0) {
4865 htmlAutoCloseOnEnd(ctxt
);
4870 * Last case, text. Note that References are handled directly.
4873 htmlParseCharData(ctxt
);
4877 if (currentNode
!= NULL
) xmlFree(currentNode
);
4882 * @ctxt: an HTML parser context
4884 * Parse a content: comment, sub-element, reference or text.
4885 * This is the entry point when called from parser.c
4889 __htmlParseContent(void *ctxt
) {
4891 htmlParseContentInternal((htmlParserCtxtPtr
) ctxt
);
4895 * htmlParseDocument:
4896 * @ctxt: an HTML parser context
4898 * parse an HTML document (and build a tree if using the standard SAX
4901 * Returns 0, -1 in case of error. the parser context is augmented
4902 * as a result of the parsing.
4906 htmlParseDocument(htmlParserCtxtPtr ctxt
) {
4908 xmlCharEncoding enc
;
4913 htmlDefaultSAXHandlerInit();
4915 if ((ctxt
== NULL
) || (ctxt
->input
== NULL
)) {
4916 htmlParseErr(ctxt
, XML_ERR_INTERNAL_ERROR
,
4917 "htmlParseDocument: context error\n", NULL
, NULL
);
4918 return(XML_ERR_INTERNAL_ERROR
);
4921 ctxt
->linenumbers
= 1;
4924 * SAX: beginning of the document processing.
4926 if ((ctxt
->sax
) && (ctxt
->sax
->setDocumentLocator
))
4927 ctxt
->sax
->setDocumentLocator(ctxt
->userData
, &xmlDefaultSAXLocator
);
4929 if ((ctxt
->encoding
== (const xmlChar
*)XML_CHAR_ENCODING_NONE
) &&
4930 ((ctxt
->input
->end
- ctxt
->input
->cur
) >= 4)) {
4932 * Get the 4 first bytes and decode the charset
4933 * if enc != XML_CHAR_ENCODING_NONE
4934 * plug some encoding conversion routines.
4940 enc
= xmlDetectCharEncoding(&start
[0], 4);
4941 if (enc
!= XML_CHAR_ENCODING_NONE
) {
4942 xmlSwitchEncoding(ctxt
, enc
);
4947 * Wipe out everything which is before the first '<'
4951 htmlParseErr(ctxt
, XML_ERR_DOCUMENT_EMPTY
,
4952 "Document is empty\n", NULL
, NULL
);
4955 if ((ctxt
->sax
) && (ctxt
->sax
->startDocument
) && (!ctxt
->disableSAX
))
4956 ctxt
->sax
->startDocument(ctxt
->userData
);
4960 * Parse possible comments and PIs before any content
4962 while (((CUR
== '<') && (NXT(1) == '!') &&
4963 (NXT(2) == '-') && (NXT(3) == '-')) ||
4964 ((CUR
== '<') && (NXT(1) == '?'))) {
4965 htmlParseComment(ctxt
);
4972 * Then possibly doc type declaration(s) and more Misc
4973 * (doctypedecl Misc*)?
4975 if ((CUR
== '<') && (NXT(1) == '!') &&
4976 (UPP(2) == 'D') && (UPP(3) == 'O') &&
4977 (UPP(4) == 'C') && (UPP(5) == 'T') &&
4978 (UPP(6) == 'Y') && (UPP(7) == 'P') &&
4980 htmlParseDocTypeDecl(ctxt
);
4985 * Parse possible comments and PIs before any content
4987 while (((CUR
== '<') && (NXT(1) == '!') &&
4988 (NXT(2) == '-') && (NXT(3) == '-')) ||
4989 ((CUR
== '<') && (NXT(1) == '?'))) {
4990 htmlParseComment(ctxt
);
4996 * Time to start parsing the tree itself
4998 htmlParseContentInternal(ctxt
);
5004 htmlAutoCloseOnEnd(ctxt
);
5008 * SAX: end of the document processing.
5010 if ((ctxt
->sax
) && (ctxt
->sax
->endDocument
!= NULL
))
5011 ctxt
->sax
->endDocument(ctxt
->userData
);
5013 if ((!(ctxt
->options
& HTML_PARSE_NODEFDTD
)) && (ctxt
->myDoc
!= NULL
)) {
5014 dtd
= xmlGetIntSubset(ctxt
->myDoc
);
5016 ctxt
->myDoc
->intSubset
=
5017 xmlCreateIntSubset(ctxt
->myDoc
, BAD_CAST
"html",
5018 BAD_CAST
"-//W3C//DTD HTML 4.0 Transitional//EN",
5019 BAD_CAST
"http://www.w3.org/TR/REC-html40/loose.dtd");
5021 if (! ctxt
->wellFormed
) return(-1);
5026 /************************************************************************
5028 * Parser contexts handling *
5030 ************************************************************************/
5033 * htmlInitParserCtxt:
5034 * @ctxt: an HTML parser context
5036 * Initialize a parser context
5038 * Returns 0 in case of success and -1 in case of error
5042 htmlInitParserCtxt(htmlParserCtxtPtr ctxt
)
5044 htmlSAXHandler
*sax
;
5046 if (ctxt
== NULL
) return(-1);
5047 memset(ctxt
, 0, sizeof(htmlParserCtxt
));
5049 ctxt
->dict
= xmlDictCreate();
5050 if (ctxt
->dict
== NULL
) {
5051 htmlErrMemory(NULL
, "htmlInitParserCtxt: out of memory\n");
5054 sax
= (htmlSAXHandler
*) xmlMalloc(sizeof(htmlSAXHandler
));
5056 htmlErrMemory(NULL
, "htmlInitParserCtxt: out of memory\n");
5059 memset(sax
, 0, sizeof(htmlSAXHandler
));
5061 /* Allocate the Input stack */
5062 ctxt
->inputTab
= (htmlParserInputPtr
*)
5063 xmlMalloc(5 * sizeof(htmlParserInputPtr
));
5064 if (ctxt
->inputTab
== NULL
) {
5065 htmlErrMemory(NULL
, "htmlInitParserCtxt: out of memory\n");
5074 ctxt
->version
= NULL
;
5075 ctxt
->encoding
= NULL
;
5076 ctxt
->standalone
= -1;
5077 ctxt
->instate
= XML_PARSER_START
;
5079 /* Allocate the Node stack */
5080 ctxt
->nodeTab
= (htmlNodePtr
*) xmlMalloc(10 * sizeof(htmlNodePtr
));
5081 if (ctxt
->nodeTab
== NULL
) {
5082 htmlErrMemory(NULL
, "htmlInitParserCtxt: out of memory\n");
5095 /* Allocate the Name stack */
5096 ctxt
->nameTab
= (const xmlChar
**) xmlMalloc(10 * sizeof(xmlChar
*));
5097 if (ctxt
->nameTab
== NULL
) {
5098 htmlErrMemory(NULL
, "htmlInitParserCtxt: out of memory\n");
5114 ctxt
->nodeInfoTab
= NULL
;
5115 ctxt
->nodeInfoNr
= 0;
5116 ctxt
->nodeInfoMax
= 0;
5119 xmlSAX2InitHtmlDefaultSAXHandler(sax
);
5121 ctxt
->userData
= ctxt
;
5123 ctxt
->wellFormed
= 1;
5124 ctxt
->replaceEntities
= 0;
5125 ctxt
->linenumbers
= xmlLineNumbersDefaultValue
;
5126 ctxt
->keepBlanks
= xmlKeepBlanksDefaultValue
;
5128 ctxt
->vctxt
.flags
= XML_VCTXT_USE_PCTXT
;
5129 ctxt
->vctxt
.userData
= ctxt
;
5130 ctxt
->vctxt
.error
= xmlParserValidityError
;
5131 ctxt
->vctxt
.warning
= xmlParserValidityWarning
;
5132 ctxt
->record_info
= 0;
5134 ctxt
->checkIndex
= 0;
5135 ctxt
->catalogs
= NULL
;
5136 xmlInitNodeInfoSeq(&ctxt
->node_seq
);
5141 * htmlFreeParserCtxt:
5142 * @ctxt: an HTML parser context
5144 * Free all the memory used by a parser context. However the parsed
5145 * document in ctxt->myDoc is not freed.
5149 htmlFreeParserCtxt(htmlParserCtxtPtr ctxt
)
5151 xmlFreeParserCtxt(ctxt
);
5155 * htmlNewParserCtxt:
5157 * Allocate and initialize a new parser context.
5159 * Returns the htmlParserCtxtPtr or NULL in case of allocation error
5163 htmlNewParserCtxt(void)
5165 xmlParserCtxtPtr ctxt
;
5167 ctxt
= (xmlParserCtxtPtr
) xmlMalloc(sizeof(xmlParserCtxt
));
5169 htmlErrMemory(NULL
, "NewParserCtxt: out of memory\n");
5172 memset(ctxt
, 0, sizeof(xmlParserCtxt
));
5173 if (htmlInitParserCtxt(ctxt
) < 0) {
5174 htmlFreeParserCtxt(ctxt
);
5181 * htmlCreateMemoryParserCtxt:
5182 * @buffer: a pointer to a char array
5183 * @size: the size of the array
5185 * Create a parser context for an HTML in-memory document.
5187 * Returns the new parser context or NULL
5190 htmlCreateMemoryParserCtxt(const char *buffer
, int size
) {
5191 xmlParserCtxtPtr ctxt
;
5192 xmlParserInputPtr input
;
5193 xmlParserInputBufferPtr buf
;
5200 ctxt
= htmlNewParserCtxt();
5204 buf
= xmlParserInputBufferCreateMem(buffer
, size
, XML_CHAR_ENCODING_NONE
);
5205 if (buf
== NULL
) return(NULL
);
5207 input
= xmlNewInputStream(ctxt
);
5208 if (input
== NULL
) {
5209 xmlFreeParserInputBuffer(buf
);
5210 xmlFreeParserCtxt(ctxt
);
5214 input
->filename
= NULL
;
5216 xmlBufResetInput(buf
->buffer
, input
);
5218 inputPush(ctxt
, input
);
5223 * htmlCreateDocParserCtxt:
5224 * @cur: a pointer to an array of xmlChar
5225 * @encoding: a free form C string describing the HTML document encoding, or NULL
5227 * Create a parser context for an HTML document.
5229 * TODO: check the need to add encoding handling there
5231 * Returns the new parser context or NULL
5233 static htmlParserCtxtPtr
5234 htmlCreateDocParserCtxt(const xmlChar
*cur
, const char *encoding
) {
5236 htmlParserCtxtPtr ctxt
;
5240 len
= xmlStrlen(cur
);
5241 ctxt
= htmlCreateMemoryParserCtxt((char *)cur
, len
);
5245 if (encoding
!= NULL
) {
5246 xmlCharEncoding enc
;
5247 xmlCharEncodingHandlerPtr handler
;
5249 if (ctxt
->input
->encoding
!= NULL
)
5250 xmlFree((xmlChar
*) ctxt
->input
->encoding
);
5251 ctxt
->input
->encoding
= xmlStrdup((const xmlChar
*) encoding
);
5253 enc
= xmlParseCharEncoding(encoding
);
5255 * registered set of known encodings
5257 if (enc
!= XML_CHAR_ENCODING_ERROR
) {
5258 xmlSwitchEncoding(ctxt
, enc
);
5259 if (ctxt
->errNo
== XML_ERR_UNSUPPORTED_ENCODING
) {
5260 htmlParseErr(ctxt
, XML_ERR_UNSUPPORTED_ENCODING
,
5261 "Unsupported encoding %s\n",
5262 (const xmlChar
*) encoding
, NULL
);
5266 * fallback for unknown encodings
5268 handler
= xmlFindCharEncodingHandler((const char *) encoding
);
5269 if (handler
!= NULL
) {
5270 xmlSwitchToEncoding(ctxt
, handler
);
5272 htmlParseErr(ctxt
, XML_ERR_UNSUPPORTED_ENCODING
,
5273 "Unsupported encoding %s\n",
5274 (const xmlChar
*) encoding
, NULL
);
5281 #ifdef LIBXML_PUSH_ENABLED
5282 /************************************************************************
5284 * Progressive parsing interfaces *
5286 ************************************************************************/
5289 * htmlParseLookupSequence:
5290 * @ctxt: an HTML parser context
5291 * @first: the first char to lookup
5292 * @next: the next char to lookup or zero
5293 * @third: the next char to lookup or zero
5294 * @ignoreattrval: skip over attribute values
5296 * Try to find if a sequence (first, next, third) or just (first next) or
5297 * (first) is available in the input stream.
5298 * This function has a side effect of (possibly) incrementing ctxt->checkIndex
5299 * to avoid rescanning sequences of bytes, it DOES change the state of the
5300 * parser, do not use liberally.
5301 * This is basically similar to xmlParseLookupSequence()
5303 * Returns the index to the current parsing point if the full sequence
5304 * is available, -1 otherwise.
5307 htmlParseLookupSequence(htmlParserCtxtPtr ctxt
, xmlChar first
,
5308 xmlChar next
, xmlChar third
, int ignoreattrval
)
5311 htmlParserInputPtr in
;
5314 char valdellim
= 0x0;
5320 base
= in
->cur
- in
->base
;
5324 if (ctxt
->checkIndex
> base
) {
5325 base
= ctxt
->checkIndex
;
5326 /* Abuse hasPErefs member to restore current state. */
5327 invalue
= ctxt
->hasPErefs
& 1 ? 1 : 0;
5330 if (in
->buf
== NULL
) {
5334 buf
= xmlBufContent(in
->buf
->buffer
);
5335 len
= xmlBufUse(in
->buf
->buffer
);
5338 /* take into account the sequence length */
5343 for (; base
< len
; base
++) {
5344 if (ignoreattrval
) {
5345 if (buf
[base
] == '"' || buf
[base
] == '\'') {
5347 if (buf
[base
] == valdellim
) {
5352 valdellim
= buf
[base
];
5356 } else if (invalue
) {
5360 if (buf
[base
] == first
) {
5362 if ((buf
[base
+ 1] != next
) || (buf
[base
+ 2] != third
))
5364 } else if (next
!= 0) {
5365 if (buf
[base
+ 1] != next
)
5368 ctxt
->checkIndex
= 0;
5371 xmlGenericError(xmlGenericErrorContext
,
5372 "HPP: lookup '%c' found at %d\n",
5374 else if (third
== 0)
5375 xmlGenericError(xmlGenericErrorContext
,
5376 "HPP: lookup '%c%c' found at %d\n",
5379 xmlGenericError(xmlGenericErrorContext
,
5380 "HPP: lookup '%c%c%c' found at %d\n",
5381 first
, next
, third
, base
);
5383 return (base
- (in
->cur
- in
->base
));
5386 ctxt
->checkIndex
= base
;
5387 /* Abuse hasPErefs member to track current state. */
5389 ctxt
->hasPErefs
|= 1;
5391 ctxt
->hasPErefs
&= ~1;
5394 xmlGenericError(xmlGenericErrorContext
,
5395 "HPP: lookup '%c' failed\n", first
);
5396 else if (third
== 0)
5397 xmlGenericError(xmlGenericErrorContext
,
5398 "HPP: lookup '%c%c' failed\n", first
, next
);
5400 xmlGenericError(xmlGenericErrorContext
,
5401 "HPP: lookup '%c%c%c' failed\n", first
, next
,
5408 * htmlParseLookupCommentEnd:
5409 * @ctxt: an HTML parser context
5411 * Try to find a comment end tag in the input stream
5412 * The search includes "-->" as well as WHATWG-recommended incorrectly-closed tags.
5413 * (See https://html.spec.whatwg.org/multipage/parsing.html#parse-error-incorrectly-closed-comment)
5414 * This function has a side effect of (possibly) incrementing ctxt->checkIndex
5415 * to avoid rescanning sequences of bytes, it DOES change the state of the
5416 * parser, do not use liberally.
5417 * This wraps to htmlParseLookupSequence()
5419 * Returns the index to the current parsing point if the full sequence is available, -1 otherwise.
5422 htmlParseLookupCommentEnd(htmlParserCtxtPtr ctxt
)
5425 int cur
= CUR_PTR
- BASE_PTR
;
5428 mark
= htmlParseLookupSequence(ctxt
, '-', '-', 0, 0);
5430 (NXT(mark
+2) == '>') ||
5431 ((NXT(mark
+2) == '!') && (NXT(mark
+3) == '>'))) {
5434 ctxt
->checkIndex
= cur
+ mark
+ 1;
5441 * htmlParseTryOrFinish:
5442 * @ctxt: an HTML parser context
5443 * @terminate: last chunk indicator
5445 * Try to progress on parsing
5447 * Returns zero if no parsing was possible
5450 htmlParseTryOrFinish(htmlParserCtxtPtr ctxt
, int terminate
) {
5452 htmlParserInputPtr in
;
5453 ptrdiff_t avail
= 0;
5456 htmlParserNodeInfo node_info
;
5459 switch (ctxt
->instate
) {
5460 case XML_PARSER_EOF
:
5461 xmlGenericError(xmlGenericErrorContext
,
5462 "HPP: try EOF\n"); break;
5463 case XML_PARSER_START
:
5464 xmlGenericError(xmlGenericErrorContext
,
5465 "HPP: try START\n"); break;
5466 case XML_PARSER_MISC
:
5467 xmlGenericError(xmlGenericErrorContext
,
5468 "HPP: try MISC\n");break;
5469 case XML_PARSER_COMMENT
:
5470 xmlGenericError(xmlGenericErrorContext
,
5471 "HPP: try COMMENT\n");break;
5472 case XML_PARSER_PROLOG
:
5473 xmlGenericError(xmlGenericErrorContext
,
5474 "HPP: try PROLOG\n");break;
5475 case XML_PARSER_START_TAG
:
5476 xmlGenericError(xmlGenericErrorContext
,
5477 "HPP: try START_TAG\n");break;
5478 case XML_PARSER_CONTENT
:
5479 xmlGenericError(xmlGenericErrorContext
,
5480 "HPP: try CONTENT\n");break;
5481 case XML_PARSER_CDATA_SECTION
:
5482 xmlGenericError(xmlGenericErrorContext
,
5483 "HPP: try CDATA_SECTION\n");break;
5484 case XML_PARSER_END_TAG
:
5485 xmlGenericError(xmlGenericErrorContext
,
5486 "HPP: try END_TAG\n");break;
5487 case XML_PARSER_ENTITY_DECL
:
5488 xmlGenericError(xmlGenericErrorContext
,
5489 "HPP: try ENTITY_DECL\n");break;
5490 case XML_PARSER_ENTITY_VALUE
:
5491 xmlGenericError(xmlGenericErrorContext
,
5492 "HPP: try ENTITY_VALUE\n");break;
5493 case XML_PARSER_ATTRIBUTE_VALUE
:
5494 xmlGenericError(xmlGenericErrorContext
,
5495 "HPP: try ATTRIBUTE_VALUE\n");break;
5496 case XML_PARSER_DTD
:
5497 xmlGenericError(xmlGenericErrorContext
,
5498 "HPP: try DTD\n");break;
5499 case XML_PARSER_EPILOG
:
5500 xmlGenericError(xmlGenericErrorContext
,
5501 "HPP: try EPILOG\n");break;
5503 xmlGenericError(xmlGenericErrorContext
,
5504 "HPP: try PI\n");break;
5505 case XML_PARSER_SYSTEM_LITERAL
:
5506 xmlGenericError(xmlGenericErrorContext
,
5507 "HPP: try SYSTEM_LITERAL\n");break;
5514 if (in
== NULL
) break;
5515 if (in
->buf
== NULL
)
5516 avail
= in
->length
- (in
->cur
- in
->base
);
5518 avail
= (ptrdiff_t)xmlBufUse(in
->buf
->buffer
) -
5519 (in
->cur
- in
->base
);
5520 if ((avail
== 0) && (terminate
)) {
5521 htmlAutoCloseOnEnd(ctxt
);
5522 if ((ctxt
->nameNr
== 0) && (ctxt
->instate
!= XML_PARSER_EOF
)) {
5524 * SAX: end of the document processing.
5526 ctxt
->instate
= XML_PARSER_EOF
;
5527 if ((ctxt
->sax
) && (ctxt
->sax
->endDocument
!= NULL
))
5528 ctxt
->sax
->endDocument(ctxt
->userData
);
5534 * This is done to make progress and avoid an infinite loop
5535 * if a parsing attempt was aborted by hitting a NUL byte. After
5536 * changing htmlCurrentChar, this probably isn't necessary anymore.
5537 * We should consider removing this check.
5545 switch (ctxt
->instate
) {
5546 case XML_PARSER_EOF
:
5548 * Document parsing is done !
5551 case XML_PARSER_START
:
5553 * Very first chars read from the document flow.
5556 if (IS_BLANK_CH(cur
)) {
5558 if (in
->buf
== NULL
)
5559 avail
= in
->length
- (in
->cur
- in
->base
);
5561 avail
= (ptrdiff_t)xmlBufUse(in
->buf
->buffer
) -
5562 (in
->cur
- in
->base
);
5564 if ((ctxt
->sax
) && (ctxt
->sax
->setDocumentLocator
))
5565 ctxt
->sax
->setDocumentLocator(ctxt
->userData
,
5566 &xmlDefaultSAXLocator
);
5567 if ((ctxt
->sax
) && (ctxt
->sax
->startDocument
) &&
5568 (!ctxt
->disableSAX
))
5569 ctxt
->sax
->startDocument(ctxt
->userData
);
5573 if ((cur
== '<') && (next
== '!') &&
5574 (UPP(2) == 'D') && (UPP(3) == 'O') &&
5575 (UPP(4) == 'C') && (UPP(5) == 'T') &&
5576 (UPP(6) == 'Y') && (UPP(7) == 'P') &&
5579 (htmlParseLookupSequence(ctxt
, '>', 0, 0, 1) < 0))
5582 xmlGenericError(xmlGenericErrorContext
,
5583 "HPP: Parsing internal subset\n");
5585 htmlParseDocTypeDecl(ctxt
);
5586 ctxt
->instate
= XML_PARSER_PROLOG
;
5588 xmlGenericError(xmlGenericErrorContext
,
5589 "HPP: entering PROLOG\n");
5592 ctxt
->instate
= XML_PARSER_MISC
;
5594 xmlGenericError(xmlGenericErrorContext
,
5595 "HPP: entering MISC\n");
5599 case XML_PARSER_MISC
:
5601 if (in
->buf
== NULL
)
5602 avail
= in
->length
- (in
->cur
- in
->base
);
5604 avail
= (ptrdiff_t)xmlBufUse(in
->buf
->buffer
) -
5605 (in
->cur
- in
->base
);
5607 * no chars in buffer
5612 * not enough chars in buffer
5623 if ((cur
== '<') && (next
== '!') &&
5624 (in
->cur
[2] == '-') && (in
->cur
[3] == '-')) {
5625 if ((!terminate
) && (htmlParseLookupCommentEnd(ctxt
) < 0))
5628 xmlGenericError(xmlGenericErrorContext
,
5629 "HPP: Parsing Comment\n");
5631 htmlParseComment(ctxt
);
5632 ctxt
->instate
= XML_PARSER_MISC
;
5633 } else if ((cur
== '<') && (next
== '?')) {
5635 (htmlParseLookupSequence(ctxt
, '>', 0, 0, 0) < 0))
5638 xmlGenericError(xmlGenericErrorContext
,
5639 "HPP: Parsing PI\n");
5642 ctxt
->instate
= XML_PARSER_MISC
;
5643 } else if ((cur
== '<') && (next
== '!') &&
5644 (UPP(2) == 'D') && (UPP(3) == 'O') &&
5645 (UPP(4) == 'C') && (UPP(5) == 'T') &&
5646 (UPP(6) == 'Y') && (UPP(7) == 'P') &&
5649 (htmlParseLookupSequence(ctxt
, '>', 0, 0, 1) < 0))
5652 xmlGenericError(xmlGenericErrorContext
,
5653 "HPP: Parsing internal subset\n");
5655 htmlParseDocTypeDecl(ctxt
);
5656 ctxt
->instate
= XML_PARSER_PROLOG
;
5658 xmlGenericError(xmlGenericErrorContext
,
5659 "HPP: entering PROLOG\n");
5661 } else if ((cur
== '<') && (next
== '!') &&
5665 ctxt
->instate
= XML_PARSER_CONTENT
;
5667 xmlGenericError(xmlGenericErrorContext
,
5668 "HPP: entering START_TAG\n");
5672 case XML_PARSER_PROLOG
:
5674 if (in
->buf
== NULL
)
5675 avail
= in
->length
- (in
->cur
- in
->base
);
5677 avail
= (ptrdiff_t)xmlBufUse(in
->buf
->buffer
) -
5678 (in
->cur
- in
->base
);
5683 if ((cur
== '<') && (next
== '!') &&
5684 (in
->cur
[2] == '-') && (in
->cur
[3] == '-')) {
5685 if ((!terminate
) && (htmlParseLookupCommentEnd(ctxt
) < 0))
5688 xmlGenericError(xmlGenericErrorContext
,
5689 "HPP: Parsing Comment\n");
5691 htmlParseComment(ctxt
);
5692 ctxt
->instate
= XML_PARSER_PROLOG
;
5693 } else if ((cur
== '<') && (next
== '?')) {
5695 (htmlParseLookupSequence(ctxt
, '>', 0, 0, 0) < 0))
5698 xmlGenericError(xmlGenericErrorContext
,
5699 "HPP: Parsing PI\n");
5702 ctxt
->instate
= XML_PARSER_PROLOG
;
5703 } else if ((cur
== '<') && (next
== '!') &&
5707 ctxt
->instate
= XML_PARSER_CONTENT
;
5709 xmlGenericError(xmlGenericErrorContext
,
5710 "HPP: entering START_TAG\n");
5714 case XML_PARSER_EPILOG
:
5715 if (in
->buf
== NULL
)
5716 avail
= in
->length
- (in
->cur
- in
->base
);
5718 avail
= (ptrdiff_t)xmlBufUse(in
->buf
->buffer
) -
5719 (in
->cur
- in
->base
);
5723 if (IS_BLANK_CH(cur
)) {
5724 htmlParseCharData(ctxt
);
5730 if ((cur
== '<') && (next
== '!') &&
5731 (in
->cur
[2] == '-') && (in
->cur
[3] == '-')) {
5732 if ((!terminate
) && (htmlParseLookupCommentEnd(ctxt
) < 0))
5735 xmlGenericError(xmlGenericErrorContext
,
5736 "HPP: Parsing Comment\n");
5738 htmlParseComment(ctxt
);
5739 ctxt
->instate
= XML_PARSER_EPILOG
;
5740 } else if ((cur
== '<') && (next
== '?')) {
5742 (htmlParseLookupSequence(ctxt
, '>', 0, 0, 0) < 0))
5745 xmlGenericError(xmlGenericErrorContext
,
5746 "HPP: Parsing PI\n");
5749 ctxt
->instate
= XML_PARSER_EPILOG
;
5750 } else if ((cur
== '<') && (next
== '!') &&
5754 ctxt
->errNo
= XML_ERR_DOCUMENT_END
;
5755 ctxt
->wellFormed
= 0;
5756 ctxt
->instate
= XML_PARSER_EOF
;
5758 xmlGenericError(xmlGenericErrorContext
,
5759 "HPP: entering EOF\n");
5761 if ((ctxt
->sax
) && (ctxt
->sax
->endDocument
!= NULL
))
5762 ctxt
->sax
->endDocument(ctxt
->userData
);
5766 case XML_PARSER_START_TAG
: {
5767 const xmlChar
*name
;
5769 const htmlElemDesc
* info
;
5772 * no chars in buffer
5777 * not enough chars in buffer
5789 ctxt
->instate
= XML_PARSER_CONTENT
;
5791 xmlGenericError(xmlGenericErrorContext
,
5792 "HPP: entering CONTENT\n");
5797 ctxt
->instate
= XML_PARSER_END_TAG
;
5798 ctxt
->checkIndex
= 0;
5800 xmlGenericError(xmlGenericErrorContext
,
5801 "HPP: entering END_TAG\n");
5806 (htmlParseLookupSequence(ctxt
, '>', 0, 0, 1) < 0))
5809 /* Capture start position */
5810 if (ctxt
->record_info
) {
5811 node_info
.begin_pos
= ctxt
->input
->consumed
+
5812 (CUR_PTR
- ctxt
->input
->base
);
5813 node_info
.begin_line
= ctxt
->input
->line
;
5817 failed
= htmlParseStartTag(ctxt
);
5819 if ((failed
== -1) ||
5827 * Lookup the info for that element.
5829 info
= htmlTagLookup(name
);
5831 htmlParseErr(ctxt
, XML_HTML_UNKNOWN_TAG
,
5832 "Tag %s invalid\n", name
, NULL
);
5836 * Check for an Empty Element labeled the XML/SGML way
5838 if ((CUR
== '/') && (NXT(1) == '>')) {
5840 if ((ctxt
->sax
!= NULL
) && (ctxt
->sax
->endElement
!= NULL
))
5841 ctxt
->sax
->endElement(ctxt
->userData
, name
);
5843 ctxt
->instate
= XML_PARSER_CONTENT
;
5845 xmlGenericError(xmlGenericErrorContext
,
5846 "HPP: entering CONTENT\n");
5854 htmlParseErr(ctxt
, XML_ERR_GT_REQUIRED
,
5855 "Couldn't find end of Start Tag %s\n",
5859 * end of parsing of this node.
5861 if (xmlStrEqual(name
, ctxt
->name
)) {
5866 if (ctxt
->record_info
)
5867 htmlNodeInfoPush(ctxt
, &node_info
);
5869 ctxt
->instate
= XML_PARSER_CONTENT
;
5871 xmlGenericError(xmlGenericErrorContext
,
5872 "HPP: entering CONTENT\n");
5878 * Check for an Empty Element from DTD definition
5880 if ((info
!= NULL
) && (info
->empty
)) {
5881 if ((ctxt
->sax
!= NULL
) && (ctxt
->sax
->endElement
!= NULL
))
5882 ctxt
->sax
->endElement(ctxt
->userData
, name
);
5886 if (ctxt
->record_info
)
5887 htmlNodeInfoPush(ctxt
, &node_info
);
5889 ctxt
->instate
= XML_PARSER_CONTENT
;
5891 xmlGenericError(xmlGenericErrorContext
,
5892 "HPP: entering CONTENT\n");
5896 case XML_PARSER_CONTENT
: {
5897 xmlChar chr
[2] = { 0, 0 };
5900 * Handle preparsed entities and charRef
5902 if (ctxt
->token
!= 0) {
5903 chr
[0] = (xmlChar
) ctxt
->token
;
5904 htmlCheckParagraph(ctxt
);
5905 if ((ctxt
->sax
!= NULL
) && (ctxt
->sax
->characters
!= NULL
))
5906 ctxt
->sax
->characters(ctxt
->userData
, chr
, 1);
5908 ctxt
->checkIndex
= 0;
5910 if ((avail
== 1) && (terminate
)) {
5912 if ((cur
!= '<') && (cur
!= '&')) {
5913 if (ctxt
->sax
!= NULL
) {
5915 if (IS_BLANK_CH(cur
)) {
5916 if (ctxt
->keepBlanks
) {
5917 if (ctxt
->sax
->characters
!= NULL
)
5918 ctxt
->sax
->characters(
5919 ctxt
->userData
, chr
, 1);
5921 if (ctxt
->sax
->ignorableWhitespace
!= NULL
)
5922 ctxt
->sax
->ignorableWhitespace(
5923 ctxt
->userData
, chr
, 1);
5926 htmlCheckParagraph(ctxt
);
5927 if (ctxt
->sax
->characters
!= NULL
)
5928 ctxt
->sax
->characters(
5929 ctxt
->userData
, chr
, 1);
5933 ctxt
->checkIndex
= 0;
5942 if ((xmlStrEqual(ctxt
->name
, BAD_CAST
"script")) ||
5943 (xmlStrEqual(ctxt
->name
, BAD_CAST
"style"))) {
5945 * Handle SCRIPT/STYLE separately
5951 idx
= htmlParseLookupSequence(ctxt
, '<', '/', 0, 0);
5954 val
= in
->cur
[idx
+ 2];
5955 if (val
== 0) /* bad cut of input */
5958 htmlParseScript(ctxt
);
5959 if ((cur
== '<') && (next
== '/')) {
5960 ctxt
->instate
= XML_PARSER_END_TAG
;
5961 ctxt
->checkIndex
= 0;
5963 xmlGenericError(xmlGenericErrorContext
,
5964 "HPP: entering END_TAG\n");
5968 } else if ((cur
== '<') && (next
== '!')) {
5970 * Sometimes DOCTYPE arrives in the middle of the document
5972 if ((UPP(2) == 'D') && (UPP(3) == 'O') &&
5973 (UPP(4) == 'C') && (UPP(5) == 'T') &&
5974 (UPP(6) == 'Y') && (UPP(7) == 'P') &&
5977 (htmlParseLookupSequence(ctxt
, '>', 0, 0, 1) < 0))
5979 htmlParseErr(ctxt
, XML_HTML_STRUCURE_ERROR
,
5980 "Misplaced DOCTYPE declaration\n",
5981 BAD_CAST
"DOCTYPE" , NULL
);
5982 htmlParseDocTypeDecl(ctxt
);
5983 } else if ((in
->cur
[2] == '-') && (in
->cur
[3] == '-')) {
5985 (htmlParseLookupCommentEnd(ctxt
) < 0))
5988 xmlGenericError(xmlGenericErrorContext
,
5989 "HPP: Parsing Comment\n");
5991 htmlParseComment(ctxt
);
5992 ctxt
->instate
= XML_PARSER_CONTENT
;
5995 (htmlParseLookupSequence(ctxt
, '>', 0, 0, 0) < 0))
5997 htmlSkipBogusComment(ctxt
);
5999 } else if ((cur
== '<') && (next
== '?')) {
6001 (htmlParseLookupSequence(ctxt
, '>', 0, 0, 0) < 0))
6004 xmlGenericError(xmlGenericErrorContext
,
6005 "HPP: Parsing PI\n");
6008 ctxt
->instate
= XML_PARSER_CONTENT
;
6009 } else if ((cur
== '<') && (next
== '!') && (avail
< 4)) {
6011 } else if ((cur
== '<') && (next
== '/')) {
6012 ctxt
->instate
= XML_PARSER_END_TAG
;
6013 ctxt
->checkIndex
= 0;
6015 xmlGenericError(xmlGenericErrorContext
,
6016 "HPP: entering END_TAG\n");
6019 } else if ((cur
== '<') && IS_ASCII_LETTER(next
)) {
6020 if ((!terminate
) && (next
== 0))
6022 ctxt
->instate
= XML_PARSER_START_TAG
;
6023 ctxt
->checkIndex
= 0;
6025 xmlGenericError(xmlGenericErrorContext
,
6026 "HPP: entering START_TAG\n");
6029 } else if (cur
== '<') {
6030 if ((ctxt
->sax
!= NULL
) && (!ctxt
->disableSAX
) &&
6031 (ctxt
->sax
->characters
!= NULL
))
6032 ctxt
->sax
->characters(ctxt
->userData
,
6037 * check that the text sequence is complete
6038 * before handing out the data to the parser
6039 * to avoid problems with erroneous end of
6043 (htmlParseLookupSequence(ctxt
, '<', 0, 0, 0) < 0))
6045 ctxt
->checkIndex
= 0;
6047 xmlGenericError(xmlGenericErrorContext
,
6048 "HPP: Parsing char data\n");
6050 while ((ctxt
->instate
!= XML_PARSER_EOF
) &&
6051 (cur
!= '<') && (in
->cur
< in
->end
)) {
6053 htmlParseReference(ctxt
);
6055 htmlParseCharData(ctxt
);
6063 case XML_PARSER_END_TAG
:
6067 (htmlParseLookupSequence(ctxt
, '>', 0, 0, 0) < 0))
6069 htmlParseEndTag(ctxt
);
6070 if (ctxt
->nameNr
== 0) {
6071 ctxt
->instate
= XML_PARSER_EPILOG
;
6073 ctxt
->instate
= XML_PARSER_CONTENT
;
6075 ctxt
->checkIndex
= 0;
6077 xmlGenericError(xmlGenericErrorContext
,
6078 "HPP: entering CONTENT\n");
6081 case XML_PARSER_CDATA_SECTION
:
6082 htmlParseErr(ctxt
, XML_ERR_INTERNAL_ERROR
,
6083 "HPP: internal error, state == CDATA\n",
6085 ctxt
->instate
= XML_PARSER_CONTENT
;
6086 ctxt
->checkIndex
= 0;
6088 xmlGenericError(xmlGenericErrorContext
,
6089 "HPP: entering CONTENT\n");
6092 case XML_PARSER_DTD
:
6093 htmlParseErr(ctxt
, XML_ERR_INTERNAL_ERROR
,
6094 "HPP: internal error, state == DTD\n",
6096 ctxt
->instate
= XML_PARSER_CONTENT
;
6097 ctxt
->checkIndex
= 0;
6099 xmlGenericError(xmlGenericErrorContext
,
6100 "HPP: entering CONTENT\n");
6103 case XML_PARSER_COMMENT
:
6104 htmlParseErr(ctxt
, XML_ERR_INTERNAL_ERROR
,
6105 "HPP: internal error, state == COMMENT\n",
6107 ctxt
->instate
= XML_PARSER_CONTENT
;
6108 ctxt
->checkIndex
= 0;
6110 xmlGenericError(xmlGenericErrorContext
,
6111 "HPP: entering CONTENT\n");
6115 htmlParseErr(ctxt
, XML_ERR_INTERNAL_ERROR
,
6116 "HPP: internal error, state == PI\n",
6118 ctxt
->instate
= XML_PARSER_CONTENT
;
6119 ctxt
->checkIndex
= 0;
6121 xmlGenericError(xmlGenericErrorContext
,
6122 "HPP: entering CONTENT\n");
6125 case XML_PARSER_ENTITY_DECL
:
6126 htmlParseErr(ctxt
, XML_ERR_INTERNAL_ERROR
,
6127 "HPP: internal error, state == ENTITY_DECL\n",
6129 ctxt
->instate
= XML_PARSER_CONTENT
;
6130 ctxt
->checkIndex
= 0;
6132 xmlGenericError(xmlGenericErrorContext
,
6133 "HPP: entering CONTENT\n");
6136 case XML_PARSER_ENTITY_VALUE
:
6137 htmlParseErr(ctxt
, XML_ERR_INTERNAL_ERROR
,
6138 "HPP: internal error, state == ENTITY_VALUE\n",
6140 ctxt
->instate
= XML_PARSER_CONTENT
;
6141 ctxt
->checkIndex
= 0;
6143 xmlGenericError(xmlGenericErrorContext
,
6144 "HPP: entering DTD\n");
6147 case XML_PARSER_ATTRIBUTE_VALUE
:
6148 htmlParseErr(ctxt
, XML_ERR_INTERNAL_ERROR
,
6149 "HPP: internal error, state == ATTRIBUTE_VALUE\n",
6151 ctxt
->instate
= XML_PARSER_START_TAG
;
6152 ctxt
->checkIndex
= 0;
6154 xmlGenericError(xmlGenericErrorContext
,
6155 "HPP: entering START_TAG\n");
6158 case XML_PARSER_SYSTEM_LITERAL
:
6159 htmlParseErr(ctxt
, XML_ERR_INTERNAL_ERROR
,
6160 "HPP: internal error, state == XML_PARSER_SYSTEM_LITERAL\n",
6162 ctxt
->instate
= XML_PARSER_CONTENT
;
6163 ctxt
->checkIndex
= 0;
6165 xmlGenericError(xmlGenericErrorContext
,
6166 "HPP: entering CONTENT\n");
6169 case XML_PARSER_IGNORE
:
6170 htmlParseErr(ctxt
, XML_ERR_INTERNAL_ERROR
,
6171 "HPP: internal error, state == XML_PARSER_IGNORE\n",
6173 ctxt
->instate
= XML_PARSER_CONTENT
;
6174 ctxt
->checkIndex
= 0;
6176 xmlGenericError(xmlGenericErrorContext
,
6177 "HPP: entering CONTENT\n");
6180 case XML_PARSER_PUBLIC_LITERAL
:
6181 htmlParseErr(ctxt
, XML_ERR_INTERNAL_ERROR
,
6182 "HPP: internal error, state == XML_PARSER_LITERAL\n",
6184 ctxt
->instate
= XML_PARSER_CONTENT
;
6185 ctxt
->checkIndex
= 0;
6187 xmlGenericError(xmlGenericErrorContext
,
6188 "HPP: entering CONTENT\n");
6195 if ((avail
== 0) && (terminate
)) {
6196 htmlAutoCloseOnEnd(ctxt
);
6197 if ((ctxt
->nameNr
== 0) && (ctxt
->instate
!= XML_PARSER_EOF
)) {
6199 * SAX: end of the document processing.
6201 ctxt
->instate
= XML_PARSER_EOF
;
6202 if ((ctxt
->sax
) && (ctxt
->sax
->endDocument
!= NULL
))
6203 ctxt
->sax
->endDocument(ctxt
->userData
);
6206 if ((!(ctxt
->options
& HTML_PARSE_NODEFDTD
)) && (ctxt
->myDoc
!= NULL
) &&
6207 ((terminate
) || (ctxt
->instate
== XML_PARSER_EOF
) ||
6208 (ctxt
->instate
== XML_PARSER_EPILOG
))) {
6210 dtd
= xmlGetIntSubset(ctxt
->myDoc
);
6212 ctxt
->myDoc
->intSubset
=
6213 xmlCreateIntSubset(ctxt
->myDoc
, BAD_CAST
"html",
6214 BAD_CAST
"-//W3C//DTD HTML 4.0 Transitional//EN",
6215 BAD_CAST
"http://www.w3.org/TR/REC-html40/loose.dtd");
6218 xmlGenericError(xmlGenericErrorContext
, "HPP: done %d\n", ret
);
6225 * @ctxt: an HTML parser context
6226 * @chunk: an char array
6227 * @size: the size in byte of the chunk
6228 * @terminate: last chunk indicator
6230 * Parse a Chunk of memory
6232 * Returns zero if no error, the xmlParserErrors otherwise.
6235 htmlParseChunk(htmlParserCtxtPtr ctxt
, const char *chunk
, int size
,
6237 if ((ctxt
== NULL
) || (ctxt
->input
== NULL
)) {
6238 htmlParseErr(ctxt
, XML_ERR_INTERNAL_ERROR
,
6239 "htmlParseChunk: context error\n", NULL
, NULL
);
6240 return(XML_ERR_INTERNAL_ERROR
);
6242 if ((size
> 0) && (chunk
!= NULL
) && (ctxt
->input
!= NULL
) &&
6243 (ctxt
->input
->buf
!= NULL
) && (ctxt
->instate
!= XML_PARSER_EOF
)) {
6244 size_t base
= xmlBufGetInputBase(ctxt
->input
->buf
->buffer
, ctxt
->input
);
6245 size_t cur
= ctxt
->input
->cur
- ctxt
->input
->base
;
6248 res
= xmlParserInputBufferPush(ctxt
->input
->buf
, size
, chunk
);
6249 xmlBufSetInputBaseCur(ctxt
->input
->buf
->buffer
, ctxt
->input
, base
, cur
);
6251 ctxt
->errNo
= XML_PARSER_EOF
;
6252 ctxt
->disableSAX
= 1;
6253 return (XML_PARSER_EOF
);
6256 xmlGenericError(xmlGenericErrorContext
, "HPP: pushed %d\n", size
);
6260 if ((terminate
) || (ctxt
->input
->buf
->buffer
->use
> 80))
6261 htmlParseTryOrFinish(ctxt
, terminate
);
6263 } else if (ctxt
->instate
!= XML_PARSER_EOF
) {
6264 if ((ctxt
->input
!= NULL
) && ctxt
->input
->buf
!= NULL
) {
6265 xmlParserInputBufferPtr in
= ctxt
->input
->buf
;
6266 if ((in
->encoder
!= NULL
) && (in
->buffer
!= NULL
) &&
6267 (in
->raw
!= NULL
)) {
6269 size_t base
= xmlBufGetInputBase(in
->buffer
, ctxt
->input
);
6270 size_t current
= ctxt
->input
->cur
- ctxt
->input
->base
;
6272 nbchars
= xmlCharEncInput(in
, terminate
);
6273 xmlBufSetInputBaseCur(in
->buffer
, ctxt
->input
, base
, current
);
6275 htmlParseErr(ctxt
, XML_ERR_INVALID_ENCODING
,
6276 "encoder error\n", NULL
, NULL
);
6277 return(XML_ERR_INVALID_ENCODING
);
6282 htmlParseTryOrFinish(ctxt
, terminate
);
6284 if ((ctxt
->instate
!= XML_PARSER_EOF
) &&
6285 (ctxt
->instate
!= XML_PARSER_EPILOG
) &&
6286 (ctxt
->instate
!= XML_PARSER_MISC
)) {
6287 ctxt
->errNo
= XML_ERR_DOCUMENT_END
;
6288 ctxt
->wellFormed
= 0;
6290 if (ctxt
->instate
!= XML_PARSER_EOF
) {
6291 if ((ctxt
->sax
) && (ctxt
->sax
->endDocument
!= NULL
))
6292 ctxt
->sax
->endDocument(ctxt
->userData
);
6294 ctxt
->instate
= XML_PARSER_EOF
;
6296 return((xmlParserErrors
) ctxt
->errNo
);
6299 /************************************************************************
6301 * User entry points *
6303 ************************************************************************/
6306 * htmlCreatePushParserCtxt:
6307 * @sax: a SAX handler
6308 * @user_data: The user data returned on SAX callbacks
6309 * @chunk: a pointer to an array of chars
6310 * @size: number of chars in the array
6311 * @filename: an optional file name or URI
6312 * @enc: an optional encoding
6314 * Create a parser context for using the HTML parser in push mode
6315 * The value of @filename is used for fetching external entities
6316 * and error/warning reports.
6318 * Returns the new parser context or NULL
6321 htmlCreatePushParserCtxt(htmlSAXHandlerPtr sax
, void *user_data
,
6322 const char *chunk
, int size
, const char *filename
,
6323 xmlCharEncoding enc
) {
6324 htmlParserCtxtPtr ctxt
;
6325 htmlParserInputPtr inputStream
;
6326 xmlParserInputBufferPtr buf
;
6330 buf
= xmlAllocParserInputBuffer(enc
);
6331 if (buf
== NULL
) return(NULL
);
6333 ctxt
= htmlNewParserCtxt();
6335 xmlFreeParserInputBuffer(buf
);
6338 if(enc
==XML_CHAR_ENCODING_UTF8
|| buf
->encoder
)
6339 ctxt
->charset
=XML_CHAR_ENCODING_UTF8
;
6341 if (ctxt
->sax
!= (xmlSAXHandlerPtr
) &htmlDefaultSAXHandler
)
6343 ctxt
->sax
= (htmlSAXHandlerPtr
) xmlMalloc(sizeof(htmlSAXHandler
));
6344 if (ctxt
->sax
== NULL
) {
6349 memcpy(ctxt
->sax
, sax
, sizeof(htmlSAXHandler
));
6350 if (user_data
!= NULL
)
6351 ctxt
->userData
= user_data
;
6353 if (filename
== NULL
) {
6354 ctxt
->directory
= NULL
;
6356 ctxt
->directory
= xmlParserGetDirectory(filename
);
6359 inputStream
= htmlNewInputStream(ctxt
);
6360 if (inputStream
== NULL
) {
6361 xmlFreeParserCtxt(ctxt
);
6366 if (filename
== NULL
)
6367 inputStream
->filename
= NULL
;
6369 inputStream
->filename
= (char *)
6370 xmlCanonicPath((const xmlChar
*) filename
);
6371 inputStream
->buf
= buf
;
6372 xmlBufResetInput(buf
->buffer
, inputStream
);
6374 inputPush(ctxt
, inputStream
);
6376 if ((size
> 0) && (chunk
!= NULL
) && (ctxt
->input
!= NULL
) &&
6377 (ctxt
->input
->buf
!= NULL
)) {
6378 size_t base
= xmlBufGetInputBase(ctxt
->input
->buf
->buffer
, ctxt
->input
);
6379 size_t cur
= ctxt
->input
->cur
- ctxt
->input
->base
;
6381 xmlParserInputBufferPush(ctxt
->input
->buf
, size
, chunk
);
6383 xmlBufSetInputBaseCur(ctxt
->input
->buf
->buffer
, ctxt
->input
, base
, cur
);
6385 xmlGenericError(xmlGenericErrorContext
, "HPP: pushed %d\n", size
);
6388 ctxt
->progressive
= 1;
6392 #endif /* LIBXML_PUSH_ENABLED */
6396 * @cur: a pointer to an array of xmlChar
6397 * @encoding: a free form C string describing the HTML document encoding, or NULL
6398 * @sax: the SAX handler block
6399 * @userData: if using SAX, this pointer will be provided on callbacks.
6401 * Parse an HTML in-memory document. If sax is not NULL, use the SAX callbacks
6402 * to handle parse events. If sax is NULL, fallback to the default DOM
6403 * behavior and return a tree.
6405 * Returns the resulting document tree unless SAX is NULL or the document is
6410 htmlSAXParseDoc(const xmlChar
*cur
, const char *encoding
,
6411 htmlSAXHandlerPtr sax
, void *userData
) {
6413 htmlParserCtxtPtr ctxt
;
6417 if (cur
== NULL
) return(NULL
);
6420 ctxt
= htmlCreateDocParserCtxt(cur
, encoding
);
6421 if (ctxt
== NULL
) return(NULL
);
6423 if (ctxt
->sax
!= NULL
) xmlFree (ctxt
->sax
);
6425 ctxt
->userData
= userData
;
6428 htmlParseDocument(ctxt
);
6432 ctxt
->userData
= NULL
;
6434 htmlFreeParserCtxt(ctxt
);
6441 * @cur: a pointer to an array of xmlChar
6442 * @encoding: a free form C string describing the HTML document encoding, or NULL
6444 * parse an HTML in-memory document and build a tree.
6446 * Returns the resulting document tree
6450 htmlParseDoc(const xmlChar
*cur
, const char *encoding
) {
6451 return(htmlSAXParseDoc(cur
, encoding
, NULL
, NULL
));
6456 * htmlCreateFileParserCtxt:
6457 * @filename: the filename
6458 * @encoding: a free form C string describing the HTML document encoding, or NULL
6460 * Create a parser context for a file content.
6461 * Automatic support for ZLIB/Compress compressed document is provided
6462 * by default if found at compile-time.
6464 * Returns the new parser context or NULL
6467 htmlCreateFileParserCtxt(const char *filename
, const char *encoding
)
6469 htmlParserCtxtPtr ctxt
;
6470 htmlParserInputPtr inputStream
;
6471 char *canonicFilename
;
6472 /* htmlCharEncoding enc; */
6473 xmlChar
*content
, *content_line
= (xmlChar
*) "charset=";
6475 if (filename
== NULL
)
6478 ctxt
= htmlNewParserCtxt();
6482 canonicFilename
= (char *) xmlCanonicPath((const xmlChar
*) filename
);
6483 if (canonicFilename
== NULL
) {
6484 #ifdef LIBXML_SAX1_ENABLED
6485 if (xmlDefaultSAXHandler
.error
!= NULL
) {
6486 xmlDefaultSAXHandler
.error(NULL
, "out of memory\n");
6489 xmlFreeParserCtxt(ctxt
);
6493 inputStream
= xmlLoadExternalEntity(canonicFilename
, NULL
, ctxt
);
6494 xmlFree(canonicFilename
);
6495 if (inputStream
== NULL
) {
6496 xmlFreeParserCtxt(ctxt
);
6500 inputPush(ctxt
, inputStream
);
6504 size_t l
= strlen(encoding
);
6507 content
= xmlMallocAtomic (xmlStrlen(content_line
) + l
+ 1);
6509 strcpy ((char *)content
, (char *)content_line
);
6510 strcat ((char *)content
, (char *)encoding
);
6511 htmlCheckEncoding (ctxt
, content
);
6522 * @filename: the filename
6523 * @encoding: a free form C string describing the HTML document encoding, or NULL
6524 * @sax: the SAX handler block
6525 * @userData: if using SAX, this pointer will be provided on callbacks.
6527 * parse an HTML file and build a tree. Automatic support for ZLIB/Compress
6528 * compressed document is provided by default if found at compile-time.
6529 * It use the given SAX function block to handle the parsing callback.
6530 * If sax is NULL, fallback to the default DOM tree building routines.
6532 * Returns the resulting document tree unless SAX is NULL or the document is
6537 htmlSAXParseFile(const char *filename
, const char *encoding
, htmlSAXHandlerPtr sax
,
6540 htmlParserCtxtPtr ctxt
;
6541 htmlSAXHandlerPtr oldsax
= NULL
;
6545 ctxt
= htmlCreateFileParserCtxt(filename
, encoding
);
6546 if (ctxt
== NULL
) return(NULL
);
6550 ctxt
->userData
= userData
;
6553 htmlParseDocument(ctxt
);
6558 ctxt
->userData
= NULL
;
6560 htmlFreeParserCtxt(ctxt
);
6567 * @filename: the filename
6568 * @encoding: a free form C string describing the HTML document encoding, or NULL
6570 * parse an HTML file and build a tree. Automatic support for ZLIB/Compress
6571 * compressed document is provided by default if found at compile-time.
6573 * Returns the resulting document tree
6577 htmlParseFile(const char *filename
, const char *encoding
) {
6578 return(htmlSAXParseFile(filename
, encoding
, NULL
, NULL
));
6582 * htmlHandleOmittedElem:
6585 * Set and return the previous value for handling HTML omitted tags.
6587 * Returns the last value for 0 for no handling, 1 for auto insertion.
6591 htmlHandleOmittedElem(int val
) {
6592 int old
= htmlOmittedDefaultValue
;
6594 htmlOmittedDefaultValue
= val
;
6599 * htmlElementAllowedHere:
6600 * @parent: HTML parent element
6601 * @elt: HTML element
6603 * Checks whether an HTML element may be a direct child of a parent element.
6604 * Note - doesn't check for deprecated elements
6606 * Returns 1 if allowed; 0 otherwise.
6609 htmlElementAllowedHere(const htmlElemDesc
* parent
, const xmlChar
* elt
) {
6612 if ( ! elt
|| ! parent
|| ! parent
->subelts
)
6615 for ( p
= parent
->subelts
; *p
; ++p
)
6616 if ( !xmlStrcmp((const xmlChar
*)*p
, elt
) )
6622 * htmlElementStatusHere:
6623 * @parent: HTML parent element
6624 * @elt: HTML element
6626 * Checks whether an HTML element may be a direct child of a parent element.
6627 * and if so whether it is valid or deprecated.
6629 * Returns one of HTML_VALID, HTML_DEPRECATED, HTML_INVALID
6632 htmlElementStatusHere(const htmlElemDesc
* parent
, const htmlElemDesc
* elt
) {
6633 if ( ! parent
|| ! elt
)
6634 return HTML_INVALID
;
6635 if ( ! htmlElementAllowedHere(parent
, (const xmlChar
*) elt
->name
) )
6636 return HTML_INVALID
;
6638 return ( elt
->dtd
== 0 ) ? HTML_VALID
: HTML_DEPRECATED
;
6642 * @elt: HTML element
6643 * @attr: HTML attribute
6644 * @legacy: whether to allow deprecated attributes
6646 * Checks whether an attribute is valid for an element
6647 * Has full knowledge of Required and Deprecated attributes
6649 * Returns one of HTML_REQUIRED, HTML_VALID, HTML_DEPRECATED, HTML_INVALID
6652 htmlAttrAllowed(const htmlElemDesc
* elt
, const xmlChar
* attr
, int legacy
) {
6655 if ( !elt
|| ! attr
)
6656 return HTML_INVALID
;
6658 if ( elt
->attrs_req
)
6659 for ( p
= elt
->attrs_req
; *p
; ++p
)
6660 if ( !xmlStrcmp((const xmlChar
*)*p
, attr
) )
6661 return HTML_REQUIRED
;
6663 if ( elt
->attrs_opt
)
6664 for ( p
= elt
->attrs_opt
; *p
; ++p
)
6665 if ( !xmlStrcmp((const xmlChar
*)*p
, attr
) )
6668 if ( legacy
&& elt
->attrs_depr
)
6669 for ( p
= elt
->attrs_depr
; *p
; ++p
)
6670 if ( !xmlStrcmp((const xmlChar
*)*p
, attr
) )
6671 return HTML_DEPRECATED
;
6673 return HTML_INVALID
;
6677 * @node: an htmlNodePtr in a tree
6678 * @legacy: whether to allow deprecated elements (YES is faster here
6679 * for Element nodes)
6681 * Checks whether the tree node is valid. Experimental (the author
6682 * only uses the HTML enhancements in a SAX parser)
6684 * Return: for Element nodes, a return from htmlElementAllowedHere (if
6685 * legacy allowed) or htmlElementStatusHere (otherwise).
6686 * for Attribute nodes, a return from htmlAttrAllowed
6687 * for other nodes, HTML_NA (no checks performed)
6690 htmlNodeStatus(const htmlNodePtr node
, int legacy
) {
6692 return HTML_INVALID
;
6694 switch ( node
->type
) {
6695 case XML_ELEMENT_NODE
:
6697 ? ( htmlElementAllowedHere (
6698 htmlTagLookup(node
->parent
->name
) , node
->name
6699 ) ? HTML_VALID
: HTML_INVALID
)
6700 : htmlElementStatusHere(
6701 htmlTagLookup(node
->parent
->name
) ,
6702 htmlTagLookup(node
->name
) )
6704 case XML_ATTRIBUTE_NODE
:
6705 return htmlAttrAllowed(
6706 htmlTagLookup(node
->parent
->name
) , node
->name
, legacy
) ;
6707 default: return HTML_NA
;
6710 /************************************************************************
6712 * New set (2.6.0) of simpler and more flexible APIs *
6714 ************************************************************************/
6719 * Free a string if it is not owned by the "dict" dictionary in the
6722 #define DICT_FREE(str) \
6723 if ((str) && ((!dict) || \
6724 (xmlDictOwns(dict, (const xmlChar *)(str)) == 0))) \
6725 xmlFree((char *)(str));
6729 * @ctxt: an HTML parser context
6731 * Reset a parser context
6734 htmlCtxtReset(htmlParserCtxtPtr ctxt
)
6736 xmlParserInputPtr input
;
6745 while ((input
= inputPop(ctxt
)) != NULL
) { /* Non consuming */
6746 xmlFreeInputStream(input
);
6752 if (ctxt
->spaceTab
!= NULL
) {
6753 ctxt
->spaceTab
[0] = -1;
6754 ctxt
->space
= &ctxt
->spaceTab
[0];
6768 DICT_FREE(ctxt
->version
);
6769 ctxt
->version
= NULL
;
6770 DICT_FREE(ctxt
->encoding
);
6771 ctxt
->encoding
= NULL
;
6772 DICT_FREE(ctxt
->directory
);
6773 ctxt
->directory
= NULL
;
6774 DICT_FREE(ctxt
->extSubURI
);
6775 ctxt
->extSubURI
= NULL
;
6776 DICT_FREE(ctxt
->extSubSystem
);
6777 ctxt
->extSubSystem
= NULL
;
6778 if (ctxt
->myDoc
!= NULL
)
6779 xmlFreeDoc(ctxt
->myDoc
);
6782 ctxt
->standalone
= -1;
6783 ctxt
->hasExternalSubset
= 0;
6784 ctxt
->hasPErefs
= 0;
6787 ctxt
->instate
= XML_PARSER_START
;
6790 ctxt
->wellFormed
= 1;
6791 ctxt
->nsWellFormed
= 1;
6792 ctxt
->disableSAX
= 0;
6794 ctxt
->vctxt
.userData
= ctxt
;
6795 ctxt
->vctxt
.error
= xmlParserValidityError
;
6796 ctxt
->vctxt
.warning
= xmlParserValidityWarning
;
6797 ctxt
->record_info
= 0;
6798 ctxt
->checkIndex
= 0;
6800 ctxt
->errNo
= XML_ERR_OK
;
6802 ctxt
->charset
= XML_CHAR_ENCODING_NONE
;
6803 ctxt
->catalogs
= NULL
;
6804 xmlInitNodeInfoSeq(&ctxt
->node_seq
);
6806 if (ctxt
->attsDefault
!= NULL
) {
6807 xmlHashFree(ctxt
->attsDefault
, xmlHashDefaultDeallocator
);
6808 ctxt
->attsDefault
= NULL
;
6810 if (ctxt
->attsSpecial
!= NULL
) {
6811 xmlHashFree(ctxt
->attsSpecial
, NULL
);
6812 ctxt
->attsSpecial
= NULL
;
6817 * htmlCtxtUseOptions:
6818 * @ctxt: an HTML parser context
6819 * @options: a combination of htmlParserOption(s)
6821 * Applies the options to the parser context
6823 * Returns 0 in case of success, the set of unknown or unimplemented options
6827 htmlCtxtUseOptions(htmlParserCtxtPtr ctxt
, int options
)
6832 if (options
& HTML_PARSE_NOWARNING
) {
6833 ctxt
->sax
->warning
= NULL
;
6834 ctxt
->vctxt
.warning
= NULL
;
6835 options
-= XML_PARSE_NOWARNING
;
6836 ctxt
->options
|= XML_PARSE_NOWARNING
;
6838 if (options
& HTML_PARSE_NOERROR
) {
6839 ctxt
->sax
->error
= NULL
;
6840 ctxt
->vctxt
.error
= NULL
;
6841 ctxt
->sax
->fatalError
= NULL
;
6842 options
-= XML_PARSE_NOERROR
;
6843 ctxt
->options
|= XML_PARSE_NOERROR
;
6845 if (options
& HTML_PARSE_PEDANTIC
) {
6847 options
-= XML_PARSE_PEDANTIC
;
6848 ctxt
->options
|= XML_PARSE_PEDANTIC
;
6851 if (options
& XML_PARSE_NOBLANKS
) {
6852 ctxt
->keepBlanks
= 0;
6853 ctxt
->sax
->ignorableWhitespace
= xmlSAX2IgnorableWhitespace
;
6854 options
-= XML_PARSE_NOBLANKS
;
6855 ctxt
->options
|= XML_PARSE_NOBLANKS
;
6857 ctxt
->keepBlanks
= 1;
6858 if (options
& HTML_PARSE_RECOVER
) {
6860 options
-= HTML_PARSE_RECOVER
;
6863 if (options
& HTML_PARSE_COMPACT
) {
6864 ctxt
->options
|= HTML_PARSE_COMPACT
;
6865 options
-= HTML_PARSE_COMPACT
;
6867 if (options
& XML_PARSE_HUGE
) {
6868 ctxt
->options
|= XML_PARSE_HUGE
;
6869 options
-= XML_PARSE_HUGE
;
6871 if (options
& HTML_PARSE_NODEFDTD
) {
6872 ctxt
->options
|= HTML_PARSE_NODEFDTD
;
6873 options
-= HTML_PARSE_NODEFDTD
;
6875 if (options
& HTML_PARSE_IGNORE_ENC
) {
6876 ctxt
->options
|= HTML_PARSE_IGNORE_ENC
;
6877 options
-= HTML_PARSE_IGNORE_ENC
;
6879 if (options
& HTML_PARSE_NOIMPLIED
) {
6880 ctxt
->options
|= HTML_PARSE_NOIMPLIED
;
6881 options
-= HTML_PARSE_NOIMPLIED
;
6883 ctxt
->dictNames
= 0;
6889 * @ctxt: an HTML parser context
6890 * @URL: the base URL to use for the document
6891 * @encoding: the document encoding, or NULL
6892 * @options: a combination of htmlParserOption(s)
6893 * @reuse: keep the context for reuse
6895 * Common front-end for the htmlRead functions
6897 * Returns the resulting document tree or NULL
6900 htmlDoRead(htmlParserCtxtPtr ctxt
, const char *URL
, const char *encoding
,
6901 int options
, int reuse
)
6905 htmlCtxtUseOptions(ctxt
, options
);
6907 if (encoding
!= NULL
) {
6908 xmlCharEncodingHandlerPtr hdlr
;
6910 hdlr
= xmlFindCharEncodingHandler(encoding
);
6912 xmlSwitchToEncoding(ctxt
, hdlr
);
6913 if (ctxt
->input
->encoding
!= NULL
)
6914 xmlFree((xmlChar
*) ctxt
->input
->encoding
);
6915 ctxt
->input
->encoding
= xmlStrdup((xmlChar
*)encoding
);
6918 if ((URL
!= NULL
) && (ctxt
->input
!= NULL
) &&
6919 (ctxt
->input
->filename
== NULL
))
6920 ctxt
->input
->filename
= (char *) xmlStrdup((const xmlChar
*) URL
);
6921 htmlParseDocument(ctxt
);
6925 if ((ctxt
->dictNames
) &&
6927 (ret
->dict
== ctxt
->dict
))
6929 xmlFreeParserCtxt(ctxt
);
6936 * @cur: a pointer to a zero terminated string
6937 * @URL: the base URL to use for the document
6938 * @encoding: the document encoding, or NULL
6939 * @options: a combination of htmlParserOption(s)
6941 * parse an XML in-memory document and build a tree.
6943 * Returns the resulting document tree
6946 htmlReadDoc(const xmlChar
* cur
, const char *URL
, const char *encoding
, int options
)
6948 htmlParserCtxtPtr ctxt
;
6954 ctxt
= htmlCreateDocParserCtxt(cur
, NULL
);
6957 return (htmlDoRead(ctxt
, URL
, encoding
, options
, 0));
6962 * @filename: a file or URL
6963 * @encoding: the document encoding, or NULL
6964 * @options: a combination of htmlParserOption(s)
6966 * parse an XML file from the filesystem or the network.
6968 * Returns the resulting document tree
6971 htmlReadFile(const char *filename
, const char *encoding
, int options
)
6973 htmlParserCtxtPtr ctxt
;
6976 ctxt
= htmlCreateFileParserCtxt(filename
, encoding
);
6979 return (htmlDoRead(ctxt
, NULL
, NULL
, options
, 0));
6984 * @buffer: a pointer to a char array
6985 * @size: the size of the array
6986 * @URL: the base URL to use for the document
6987 * @encoding: the document encoding, or NULL
6988 * @options: a combination of htmlParserOption(s)
6990 * parse an XML in-memory document and build a tree.
6992 * Returns the resulting document tree
6995 htmlReadMemory(const char *buffer
, int size
, const char *URL
, const char *encoding
, int options
)
6997 htmlParserCtxtPtr ctxt
;
7000 ctxt
= xmlCreateMemoryParserCtxt(buffer
, size
);
7003 htmlDefaultSAXHandlerInit();
7004 if (ctxt
->sax
!= NULL
)
7005 memcpy(ctxt
->sax
, &htmlDefaultSAXHandler
, sizeof(xmlSAXHandlerV1
));
7006 return (htmlDoRead(ctxt
, URL
, encoding
, options
, 0));
7011 * @fd: an open file descriptor
7012 * @URL: the base URL to use for the document
7013 * @encoding: the document encoding, or NULL
7014 * @options: a combination of htmlParserOption(s)
7016 * parse an HTML from a file descriptor and build a tree.
7017 * NOTE that the file descriptor will not be closed when the
7018 * reader is closed or reset.
7020 * Returns the resulting document tree
7023 htmlReadFd(int fd
, const char *URL
, const char *encoding
, int options
)
7025 htmlParserCtxtPtr ctxt
;
7026 xmlParserInputBufferPtr input
;
7027 htmlParserInputPtr stream
;
7033 input
= xmlParserInputBufferCreateFd(fd
, XML_CHAR_ENCODING_NONE
);
7036 input
->closecallback
= NULL
;
7037 ctxt
= htmlNewParserCtxt();
7039 xmlFreeParserInputBuffer(input
);
7042 stream
= xmlNewIOInputStream(ctxt
, input
, XML_CHAR_ENCODING_NONE
);
7043 if (stream
== NULL
) {
7044 xmlFreeParserInputBuffer(input
);
7045 htmlFreeParserCtxt(ctxt
);
7048 inputPush(ctxt
, stream
);
7049 return (htmlDoRead(ctxt
, URL
, encoding
, options
, 0));
7054 * @ioread: an I/O read function
7055 * @ioclose: an I/O close function
7056 * @ioctx: an I/O handler
7057 * @URL: the base URL to use for the document
7058 * @encoding: the document encoding, or NULL
7059 * @options: a combination of htmlParserOption(s)
7061 * parse an HTML document from I/O functions and source and build a tree.
7063 * Returns the resulting document tree
7066 htmlReadIO(xmlInputReadCallback ioread
, xmlInputCloseCallback ioclose
,
7067 void *ioctx
, const char *URL
, const char *encoding
, int options
)
7069 htmlParserCtxtPtr ctxt
;
7070 xmlParserInputBufferPtr input
;
7071 xmlParserInputPtr stream
;
7077 input
= xmlParserInputBufferCreateIO(ioread
, ioclose
, ioctx
,
7078 XML_CHAR_ENCODING_NONE
);
7079 if (input
== NULL
) {
7080 if (ioclose
!= NULL
)
7084 ctxt
= htmlNewParserCtxt();
7086 xmlFreeParserInputBuffer(input
);
7089 stream
= xmlNewIOInputStream(ctxt
, input
, XML_CHAR_ENCODING_NONE
);
7090 if (stream
== NULL
) {
7091 xmlFreeParserInputBuffer(input
);
7092 xmlFreeParserCtxt(ctxt
);
7095 inputPush(ctxt
, stream
);
7096 return (htmlDoRead(ctxt
, URL
, encoding
, options
, 0));
7101 * @ctxt: an HTML parser context
7102 * @cur: a pointer to a zero terminated string
7103 * @URL: the base URL to use for the document
7104 * @encoding: the document encoding, or NULL
7105 * @options: a combination of htmlParserOption(s)
7107 * parse an XML in-memory document and build a tree.
7108 * This reuses the existing @ctxt parser context
7110 * Returns the resulting document tree
7113 htmlCtxtReadDoc(htmlParserCtxtPtr ctxt
, const xmlChar
* cur
,
7114 const char *URL
, const char *encoding
, int options
)
7118 return (htmlCtxtReadMemory(ctxt
, (const char *) cur
, xmlStrlen(cur
), URL
,
7119 encoding
, options
));
7124 * @ctxt: an HTML parser context
7125 * @filename: a file or URL
7126 * @encoding: the document encoding, or NULL
7127 * @options: a combination of htmlParserOption(s)
7129 * parse an XML file from the filesystem or the network.
7130 * This reuses the existing @ctxt parser context
7132 * Returns the resulting document tree
7135 htmlCtxtReadFile(htmlParserCtxtPtr ctxt
, const char *filename
,
7136 const char *encoding
, int options
)
7138 xmlParserInputPtr stream
;
7140 if (filename
== NULL
)
7146 htmlCtxtReset(ctxt
);
7148 stream
= xmlLoadExternalEntity(filename
, NULL
, ctxt
);
7149 if (stream
== NULL
) {
7152 inputPush(ctxt
, stream
);
7153 return (htmlDoRead(ctxt
, NULL
, encoding
, options
, 1));
7157 * htmlCtxtReadMemory:
7158 * @ctxt: an HTML parser context
7159 * @buffer: a pointer to a char array
7160 * @size: the size of the array
7161 * @URL: the base URL to use for the document
7162 * @encoding: the document encoding, or NULL
7163 * @options: a combination of htmlParserOption(s)
7165 * parse an XML in-memory document and build a tree.
7166 * This reuses the existing @ctxt parser context
7168 * Returns the resulting document tree
7171 htmlCtxtReadMemory(htmlParserCtxtPtr ctxt
, const char *buffer
, int size
,
7172 const char *URL
, const char *encoding
, int options
)
7174 xmlParserInputBufferPtr input
;
7175 xmlParserInputPtr stream
;
7183 htmlCtxtReset(ctxt
);
7185 input
= xmlParserInputBufferCreateMem(buffer
, size
, XML_CHAR_ENCODING_NONE
);
7186 if (input
== NULL
) {
7190 stream
= xmlNewIOInputStream(ctxt
, input
, XML_CHAR_ENCODING_NONE
);
7191 if (stream
== NULL
) {
7192 xmlFreeParserInputBuffer(input
);
7196 inputPush(ctxt
, stream
);
7197 return (htmlDoRead(ctxt
, URL
, encoding
, options
, 1));
7202 * @ctxt: an HTML parser context
7203 * @fd: an open file descriptor
7204 * @URL: the base URL to use for the document
7205 * @encoding: the document encoding, or NULL
7206 * @options: a combination of htmlParserOption(s)
7208 * parse an XML from a file descriptor and build a tree.
7209 * This reuses the existing @ctxt parser context
7211 * Returns the resulting document tree
7214 htmlCtxtReadFd(htmlParserCtxtPtr ctxt
, int fd
,
7215 const char *URL
, const char *encoding
, int options
)
7217 xmlParserInputBufferPtr input
;
7218 xmlParserInputPtr stream
;
7226 htmlCtxtReset(ctxt
);
7229 input
= xmlParserInputBufferCreateFd(fd
, XML_CHAR_ENCODING_NONE
);
7232 stream
= xmlNewIOInputStream(ctxt
, input
, XML_CHAR_ENCODING_NONE
);
7233 if (stream
== NULL
) {
7234 xmlFreeParserInputBuffer(input
);
7237 inputPush(ctxt
, stream
);
7238 return (htmlDoRead(ctxt
, URL
, encoding
, options
, 1));
7243 * @ctxt: an HTML parser context
7244 * @ioread: an I/O read function
7245 * @ioclose: an I/O close function
7246 * @ioctx: an I/O handler
7247 * @URL: the base URL to use for the document
7248 * @encoding: the document encoding, or NULL
7249 * @options: a combination of htmlParserOption(s)
7251 * parse an HTML document from I/O functions and source and build a tree.
7252 * This reuses the existing @ctxt parser context
7254 * Returns the resulting document tree
7257 htmlCtxtReadIO(htmlParserCtxtPtr ctxt
, xmlInputReadCallback ioread
,
7258 xmlInputCloseCallback ioclose
, void *ioctx
,
7260 const char *encoding
, int options
)
7262 xmlParserInputBufferPtr input
;
7263 xmlParserInputPtr stream
;
7271 htmlCtxtReset(ctxt
);
7273 input
= xmlParserInputBufferCreateIO(ioread
, ioclose
, ioctx
,
7274 XML_CHAR_ENCODING_NONE
);
7275 if (input
== NULL
) {
7276 if (ioclose
!= NULL
)
7280 stream
= xmlNewIOInputStream(ctxt
, input
, XML_CHAR_ENCODING_NONE
);
7281 if (stream
== NULL
) {
7282 xmlFreeParserInputBuffer(input
);
7285 inputPush(ctxt
, stream
);
7286 return (htmlDoRead(ctxt
, URL
, encoding
, options
, 1));
7289 #endif /* LIBXML_HTML_ENABLED */