2 * HTMLtree.c : implementation of access function for an HTML tree.
4 * See Copyright for the status of this software.
12 #ifdef LIBXML_HTML_ENABLED
14 #include <string.h> /* for memset() only ! */
23 #include <libxml/xmlmemory.h>
24 #include <libxml/HTMLparser.h>
25 #include <libxml/HTMLtree.h>
26 #include <libxml/entities.h>
27 #include <libxml/valid.h>
28 #include <libxml/xmlerror.h>
29 #include <libxml/parserInternals.h>
30 #include <libxml/globals.h>
31 #include <libxml/uri.h>
35 /************************************************************************
37 * Getting/Setting encoding meta tags *
39 ************************************************************************/
42 * htmlGetMetaEncoding:
45 * Encoding definition lookup in the Meta tags
47 * Returns the current encoding as flagged in the HTML source
50 htmlGetMetaEncoding(htmlDocPtr doc
) {
52 const xmlChar
*content
;
53 const xmlChar
*encoding
;
63 if ((cur
->type
== XML_ELEMENT_NODE
) && (cur
->name
!= NULL
)) {
64 if (xmlStrEqual(cur
->name
, BAD_CAST
"html"))
66 if (xmlStrEqual(cur
->name
, BAD_CAST
"head"))
68 if (xmlStrEqual(cur
->name
, BAD_CAST
"meta"))
81 if ((cur
->type
== XML_ELEMENT_NODE
) && (cur
->name
!= NULL
)) {
82 if (xmlStrEqual(cur
->name
, BAD_CAST
"head"))
84 if (xmlStrEqual(cur
->name
, BAD_CAST
"meta"))
95 * Search the meta elements
99 if ((cur
->type
== XML_ELEMENT_NODE
) && (cur
->name
!= NULL
)) {
100 if (xmlStrEqual(cur
->name
, BAD_CAST
"meta")) {
101 xmlAttrPtr attr
= cur
->properties
;
103 const xmlChar
*value
;
107 while (attr
!= NULL
) {
108 if ((attr
->children
!= NULL
) &&
109 (attr
->children
->type
== XML_TEXT_NODE
) &&
110 (attr
->children
->next
== NULL
)) {
111 value
= attr
->children
->content
;
112 if ((!xmlStrcasecmp(attr
->name
, BAD_CAST
"http-equiv"))
113 && (!xmlStrcasecmp(value
, BAD_CAST
"Content-Type")))
115 else if ((value
!= NULL
)
116 && (!xmlStrcasecmp(attr
->name
, BAD_CAST
"content")))
118 if ((http
!= 0) && (content
!= NULL
))
130 encoding
= xmlStrstr(content
, BAD_CAST
"charset=");
131 if (encoding
== NULL
)
132 encoding
= xmlStrstr(content
, BAD_CAST
"Charset=");
133 if (encoding
== NULL
)
134 encoding
= xmlStrstr(content
, BAD_CAST
"CHARSET=");
135 if (encoding
!= NULL
) {
138 encoding
= xmlStrstr(content
, BAD_CAST
"charset =");
139 if (encoding
== NULL
)
140 encoding
= xmlStrstr(content
, BAD_CAST
"Charset =");
141 if (encoding
== NULL
)
142 encoding
= xmlStrstr(content
, BAD_CAST
"CHARSET =");
143 if (encoding
!= NULL
)
146 if (encoding
!= NULL
) {
147 while ((*encoding
== ' ') || (*encoding
== '\t')) encoding
++;
153 * htmlSetMetaEncoding:
155 * @encoding: the encoding string
157 * Sets the current encoding in the Meta tags
158 * NOTE: this will not change the document content encoding, just
159 * the META flag associated.
161 * Returns 0 in case of success and -1 in case of error
164 htmlSetMetaEncoding(htmlDocPtr doc
, const xmlChar
*encoding
) {
165 htmlNodePtr cur
, meta
= NULL
, head
= NULL
;
166 const xmlChar
*content
= NULL
;
167 char newcontent
[100];
174 /* html isn't a real encoding it's just libxml2 way to get entities */
175 if (!xmlStrcasecmp(encoding
, BAD_CAST
"html"))
178 if (encoding
!= NULL
) {
179 snprintf(newcontent
, sizeof(newcontent
), "text/html; charset=%s",
181 newcontent
[sizeof(newcontent
) - 1] = 0;
189 while (cur
!= NULL
) {
190 if ((cur
->type
== XML_ELEMENT_NODE
) && (cur
->name
!= NULL
)) {
191 if (xmlStrcasecmp(cur
->name
, BAD_CAST
"html") == 0)
193 if (xmlStrcasecmp(cur
->name
, BAD_CAST
"head") == 0)
195 if (xmlStrcasecmp(cur
->name
, BAD_CAST
"meta") == 0)
207 while (cur
!= NULL
) {
208 if ((cur
->type
== XML_ELEMENT_NODE
) && (cur
->name
!= NULL
)) {
209 if (xmlStrcasecmp(cur
->name
, BAD_CAST
"head") == 0)
211 if (xmlStrcasecmp(cur
->name
, BAD_CAST
"meta") == 0) {
222 if (cur
->children
== NULL
)
228 * Search and update all the remaining the meta elements carrying
229 * encoding informations
231 while (cur
!= NULL
) {
232 if ((cur
->type
== XML_ELEMENT_NODE
) && (cur
->name
!= NULL
)) {
233 if (xmlStrcasecmp(cur
->name
, BAD_CAST
"meta") == 0) {
234 xmlAttrPtr attr
= cur
->properties
;
236 const xmlChar
*value
;
240 while (attr
!= NULL
) {
241 if ((attr
->children
!= NULL
) &&
242 (attr
->children
->type
== XML_TEXT_NODE
) &&
243 (attr
->children
->next
== NULL
)) {
244 value
= attr
->children
->content
;
245 if ((!xmlStrcasecmp(attr
->name
, BAD_CAST
"http-equiv"))
246 && (!xmlStrcasecmp(value
, BAD_CAST
"Content-Type")))
250 if ((value
!= NULL
) &&
251 (!xmlStrcasecmp(attr
->name
, BAD_CAST
"content")))
254 if ((http
!= 0) && (content
!= NULL
))
259 if ((http
!= 0) && (content
!= NULL
)) {
270 if ((encoding
!= NULL
) && (head
!= NULL
)) {
272 * Create a new Meta element with the right attributes
275 meta
= xmlNewDocNode(doc
, NULL
, BAD_CAST
"meta", NULL
);
276 if (head
->children
== NULL
)
277 xmlAddChild(head
, meta
);
279 xmlAddPrevSibling(head
->children
, meta
);
280 xmlNewProp(meta
, BAD_CAST
"http-equiv", BAD_CAST
"Content-Type");
281 xmlNewProp(meta
, BAD_CAST
"content", BAD_CAST newcontent
);
284 /* remove the meta tag if NULL is passed */
285 if (encoding
== NULL
) {
289 /* change the document only if there is a real encoding change */
290 else if (xmlStrcasestr(content
, encoding
) == NULL
) {
291 xmlSetProp(meta
, BAD_CAST
"content", BAD_CAST newcontent
);
302 * These are the HTML attributes which will be output
303 * in minimized form, i.e. <option selected="selected"> will be
304 * output as <option selected>, as per XSLT 1.0 16.2 "HTML Output Method"
307 static const char* htmlBooleanAttrs
[] = {
308 "checked", "compact", "declare", "defer", "disabled", "ismap",
309 "multiple", "nohref", "noresize", "noshade", "nowrap", "readonly",
316 * @name: the name of the attribute to check
318 * Determine if a given attribute is a boolean attribute.
320 * returns: false if the attribute is not boolean, true otherwise.
323 htmlIsBooleanAttr(const xmlChar
*name
)
327 while (htmlBooleanAttrs
[i
] != NULL
) {
328 if (xmlStrcasecmp((const xmlChar
*)htmlBooleanAttrs
[i
], name
) == 0)
335 #ifdef LIBXML_OUTPUT_ENABLED
337 * private routine exported from xmlIO.c
340 xmlAllocOutputBufferInternal(xmlCharEncodingHandlerPtr encoder
);
341 /************************************************************************
343 * Output error handlers *
345 ************************************************************************/
348 * @extra: extra informations
350 * Handle an out of memory condition
353 htmlSaveErrMemory(const char *extra
)
355 __xmlSimpleError(XML_FROM_OUTPUT
, XML_ERR_NO_MEMORY
, NULL
, NULL
, extra
);
360 * @code: the error number
361 * @node: the location of the error.
362 * @extra: extra informations
364 * Handle an out of memory condition
367 htmlSaveErr(int code
, xmlNodePtr node
, const char *extra
)
369 const char *msg
= NULL
;
372 case XML_SAVE_NOT_UTF8
:
373 msg
= "string is not in UTF-8\n";
375 case XML_SAVE_CHAR_INVALID
:
376 msg
= "invalid character value\n";
378 case XML_SAVE_UNKNOWN_ENCODING
:
379 msg
= "unknown encoding %s\n";
381 case XML_SAVE_NO_DOCTYPE
:
382 msg
= "HTML has no DOCTYPE\n";
385 msg
= "unexpected error number\n";
387 __xmlSimpleError(XML_FROM_OUTPUT
, code
, node
, msg
, extra
);
390 /************************************************************************
392 * Dumping HTML tree content to a simple buffer *
394 ************************************************************************/
397 * htmlBufNodeDumpFormat:
398 * @buf: the xmlBufPtr output
400 * @cur: the current node
401 * @format: should formatting spaces been added
403 * Dump an HTML node, recursive behaviour,children are printed too.
405 * Returns the number of byte written or -1 in case of error
408 htmlBufNodeDumpFormat(xmlBufPtr buf
, xmlDocPtr doc
, xmlNodePtr cur
,
412 xmlOutputBufferPtr outbuf
;
420 outbuf
= (xmlOutputBufferPtr
) xmlMalloc(sizeof(xmlOutputBuffer
));
421 if (outbuf
== NULL
) {
422 htmlSaveErrMemory("allocating HTML output buffer");
425 memset(outbuf
, 0, (size_t) sizeof(xmlOutputBuffer
));
426 outbuf
->buffer
= buf
;
427 outbuf
->encoder
= NULL
;
428 outbuf
->writecallback
= NULL
;
429 outbuf
->closecallback
= NULL
;
430 outbuf
->context
= NULL
;
433 use
= xmlBufUse(buf
);
434 htmlNodeDumpFormatOutput(outbuf
, doc
, cur
, NULL
, format
);
436 ret
= xmlBufUse(buf
) - use
;
442 * @buf: the HTML buffer output
444 * @cur: the current node
446 * Dump an HTML node, recursive behaviour,children are printed too,
447 * and formatting returns are added.
449 * Returns the number of byte written or -1 in case of error
452 htmlNodeDump(xmlBufferPtr buf
, xmlDocPtr doc
, xmlNodePtr cur
) {
456 if ((buf
== NULL
) || (cur
== NULL
))
460 buffer
= xmlBufFromBuffer(buf
);
464 ret
= htmlBufNodeDumpFormat(buffer
, doc
, cur
, 1);
466 xmlBufBackToBuffer(buffer
);
474 * htmlNodeDumpFileFormat:
475 * @out: the FILE pointer
477 * @cur: the current node
478 * @encoding: the document encoding
479 * @format: should formatting spaces been added
481 * Dump an HTML node, recursive behaviour,children are printed too.
483 * TODO: if encoding == NULL try to save in the doc encoding
485 * returns: the number of byte written or -1 in case of failure.
488 htmlNodeDumpFileFormat(FILE *out
, xmlDocPtr doc
,
489 xmlNodePtr cur
, const char *encoding
, int format
) {
490 xmlOutputBufferPtr buf
;
491 xmlCharEncodingHandlerPtr handler
= NULL
;
496 if (encoding
!= NULL
) {
499 enc
= xmlParseCharEncoding(encoding
);
500 if (enc
!= XML_CHAR_ENCODING_UTF8
) {
501 handler
= xmlFindCharEncodingHandler(encoding
);
503 htmlSaveErr(XML_SAVE_UNKNOWN_ENCODING
, NULL
, encoding
);
508 * Fallback to HTML or ASCII when the encoding is unspecified
511 handler
= xmlFindCharEncodingHandler("HTML");
513 handler
= xmlFindCharEncodingHandler("ascii");
516 * save the content to a temp buffer.
518 buf
= xmlOutputBufferCreateFile(out
, handler
);
519 if (buf
== NULL
) return(0);
521 htmlNodeDumpFormatOutput(buf
, doc
, cur
, encoding
, format
);
523 ret
= xmlOutputBufferClose(buf
);
529 * @out: the FILE pointer
531 * @cur: the current node
533 * Dump an HTML node, recursive behaviour,children are printed too,
534 * and formatting returns are added.
537 htmlNodeDumpFile(FILE *out
, xmlDocPtr doc
, xmlNodePtr cur
) {
538 htmlNodeDumpFileFormat(out
, doc
, cur
, NULL
, 1);
542 * htmlDocDumpMemoryFormat:
544 * @mem: OUT: the memory pointer
545 * @size: OUT: the memory length
546 * @format: should formatting spaces been added
548 * Dump an HTML document in memory and return the xmlChar * and it's size.
549 * It's up to the caller to free the memory.
552 htmlDocDumpMemoryFormat(xmlDocPtr cur
, xmlChar
**mem
, int *size
, int format
) {
553 xmlOutputBufferPtr buf
;
554 xmlCharEncodingHandlerPtr handler
= NULL
;
555 const char *encoding
;
559 if ((mem
== NULL
) || (size
== NULL
))
567 encoding
= (const char *) htmlGetMetaEncoding(cur
);
569 if (encoding
!= NULL
) {
572 enc
= xmlParseCharEncoding(encoding
);
573 if (enc
!= cur
->charset
) {
574 if (cur
->charset
!= XML_CHAR_ENCODING_UTF8
) {
583 handler
= xmlFindCharEncodingHandler(encoding
);
585 htmlSaveErr(XML_SAVE_UNKNOWN_ENCODING
, NULL
, encoding
);
588 handler
= xmlFindCharEncodingHandler(encoding
);
593 * Fallback to HTML or ASCII when the encoding is unspecified
596 handler
= xmlFindCharEncodingHandler("HTML");
598 handler
= xmlFindCharEncodingHandler("ascii");
600 buf
= xmlAllocOutputBufferInternal(handler
);
607 htmlDocContentDumpFormatOutput(buf
, cur
, NULL
, format
);
609 xmlOutputBufferFlush(buf
);
610 if (buf
->conv
!= NULL
) {
611 *size
= xmlBufUse(buf
->conv
);
612 *mem
= xmlStrndup(xmlBufContent(buf
->conv
), *size
);
614 *size
= xmlBufUse(buf
->buffer
);
615 *mem
= xmlStrndup(xmlBufContent(buf
->buffer
), *size
);
617 (void)xmlOutputBufferClose(buf
);
623 * @mem: OUT: the memory pointer
624 * @size: OUT: the memory length
626 * Dump an HTML document in memory and return the xmlChar * and it's size.
627 * It's up to the caller to free the memory.
630 htmlDocDumpMemory(xmlDocPtr cur
, xmlChar
**mem
, int *size
) {
631 htmlDocDumpMemoryFormat(cur
, mem
, size
, 1);
635 /************************************************************************
637 * Dumping HTML tree content to an I/O output buffer *
639 ************************************************************************/
641 void xmlNsListDumpOutput(xmlOutputBufferPtr buf
, xmlNsPtr cur
);
645 * @buf: the HTML buffer output
647 * @encoding: the encoding string
649 * TODO: check whether encoding is needed
651 * Dump the HTML document DTD, if any.
654 htmlDtdDumpOutput(xmlOutputBufferPtr buf
, xmlDocPtr doc
,
655 const char *encoding ATTRIBUTE_UNUSED
) {
656 xmlDtdPtr cur
= doc
->intSubset
;
659 htmlSaveErr(XML_SAVE_NO_DOCTYPE
, (xmlNodePtr
) doc
, NULL
);
662 xmlOutputBufferWriteString(buf
, "<!DOCTYPE ");
663 xmlOutputBufferWriteString(buf
, (const char *)cur
->name
);
664 if (cur
->ExternalID
!= NULL
) {
665 xmlOutputBufferWriteString(buf
, " PUBLIC ");
666 xmlBufWriteQuotedString(buf
->buffer
, cur
->ExternalID
);
667 if (cur
->SystemID
!= NULL
) {
668 xmlOutputBufferWriteString(buf
, " ");
669 xmlBufWriteQuotedString(buf
->buffer
, cur
->SystemID
);
671 } else if (cur
->SystemID
!= NULL
) {
672 xmlOutputBufferWriteString(buf
, " SYSTEM ");
673 xmlBufWriteQuotedString(buf
->buffer
, cur
->SystemID
);
675 xmlOutputBufferWriteString(buf
, ">\n");
679 * htmlAttrDumpOutput:
680 * @buf: the HTML buffer output
682 * @cur: the attribute pointer
683 * @encoding: the encoding string
685 * Dump an HTML attribute
688 htmlAttrDumpOutput(xmlOutputBufferPtr buf
, xmlDocPtr doc
, xmlAttrPtr cur
,
689 const char *encoding ATTRIBUTE_UNUSED
) {
693 * The html output method should not escape a & character
694 * occurring in an attribute value immediately followed by
695 * a { character (see Section B.7.1 of the HTML 4.0 Recommendation).
696 * This is implemented in xmlEncodeEntitiesReentrant
702 xmlOutputBufferWriteString(buf
, " ");
703 if ((cur
->ns
!= NULL
) && (cur
->ns
->prefix
!= NULL
)) {
704 xmlOutputBufferWriteString(buf
, (const char *)cur
->ns
->prefix
);
705 xmlOutputBufferWriteString(buf
, ":");
707 xmlOutputBufferWriteString(buf
, (const char *)cur
->name
);
708 if ((cur
->children
!= NULL
) && (!htmlIsBooleanAttr(cur
->name
))) {
709 value
= xmlNodeListGetString(doc
, cur
->children
, 0);
711 xmlOutputBufferWriteString(buf
, "=");
712 if ((cur
->ns
== NULL
) && (cur
->parent
!= NULL
) &&
713 (cur
->parent
->ns
== NULL
) &&
714 ((!xmlStrcasecmp(cur
->name
, BAD_CAST
"href")) ||
715 (!xmlStrcasecmp(cur
->name
, BAD_CAST
"action")) ||
716 (!xmlStrcasecmp(cur
->name
, BAD_CAST
"src")) ||
717 ((!xmlStrcasecmp(cur
->name
, BAD_CAST
"name")) &&
718 (!xmlStrcasecmp(cur
->parent
->name
, BAD_CAST
"a"))))) {
720 xmlChar
*tmp
= value
;
722 while (IS_BLANK_CH(*tmp
)) tmp
++;
725 * the < and > have already been escaped at the entity level
726 * And doing so here breaks server side includes
728 escaped
= xmlURIEscapeStr(tmp
, BAD_CAST
"@/:=?;#%&,+<>");
729 if (escaped
!= NULL
) {
730 xmlBufWriteQuotedString(buf
->buffer
, escaped
);
733 xmlBufWriteQuotedString(buf
->buffer
, value
);
736 xmlBufWriteQuotedString(buf
->buffer
, value
);
740 xmlOutputBufferWriteString(buf
, "=\"\"");
746 * htmlAttrListDumpOutput:
747 * @buf: the HTML buffer output
749 * @cur: the first attribute pointer
750 * @encoding: the encoding string
752 * Dump a list of HTML attributes
755 htmlAttrListDumpOutput(xmlOutputBufferPtr buf
, xmlDocPtr doc
, xmlAttrPtr cur
, const char *encoding
) {
759 while (cur
!= NULL
) {
760 htmlAttrDumpOutput(buf
, doc
, cur
, encoding
);
768 * htmlNodeListDumpOutput:
769 * @buf: the HTML buffer output
771 * @cur: the first node
772 * @encoding: the encoding string
773 * @format: should formatting spaces been added
775 * Dump an HTML node list, recursive behaviour,children are printed too.
778 htmlNodeListDumpOutput(xmlOutputBufferPtr buf
, xmlDocPtr doc
,
779 xmlNodePtr cur
, const char *encoding
, int format
) {
783 while (cur
!= NULL
) {
784 htmlNodeDumpFormatOutput(buf
, doc
, cur
, encoding
, format
);
790 * htmlNodeDumpFormatOutput:
791 * @buf: the HTML buffer output
793 * @cur: the current node
794 * @encoding: the encoding string
795 * @format: should formatting spaces been added
797 * Dump an HTML node, recursive behaviour,children are printed too.
800 htmlNodeDumpFormatOutput(xmlOutputBufferPtr buf
, xmlDocPtr doc
,
801 xmlNodePtr cur
, const char *encoding
, int format
) {
802 const htmlElemDesc
* info
;
806 if ((cur
== NULL
) || (buf
== NULL
)) {
812 if (cur
->type
== XML_DTD_NODE
)
814 if ((cur
->type
== XML_HTML_DOCUMENT_NODE
) ||
815 (cur
->type
== XML_DOCUMENT_NODE
)){
816 htmlDocContentDumpOutput(buf
, (xmlDocPtr
) cur
, encoding
);
819 if (cur
->type
== XML_ATTRIBUTE_NODE
) {
820 htmlAttrDumpOutput(buf
, doc
, (xmlAttrPtr
) cur
, encoding
);
823 if (cur
->type
== HTML_TEXT_NODE
) {
824 if (cur
->content
!= NULL
) {
825 if (((cur
->name
== (const xmlChar
*)xmlStringText
) ||
826 (cur
->name
!= (const xmlChar
*)xmlStringTextNoenc
)) &&
827 ((cur
->parent
== NULL
) ||
828 ((xmlStrcasecmp(cur
->parent
->name
, BAD_CAST
"script")) &&
829 (xmlStrcasecmp(cur
->parent
->name
, BAD_CAST
"style"))))) {
832 buffer
= xmlEncodeEntitiesReentrant(doc
, cur
->content
);
833 if (buffer
!= NULL
) {
834 xmlOutputBufferWriteString(buf
, (const char *)buffer
);
838 xmlOutputBufferWriteString(buf
, (const char *)cur
->content
);
843 if (cur
->type
== HTML_COMMENT_NODE
) {
844 if (cur
->content
!= NULL
) {
845 xmlOutputBufferWriteString(buf
, "<!--");
846 xmlOutputBufferWriteString(buf
, (const char *)cur
->content
);
847 xmlOutputBufferWriteString(buf
, "-->");
851 if (cur
->type
== HTML_PI_NODE
) {
852 if (cur
->name
== NULL
)
854 xmlOutputBufferWriteString(buf
, "<?");
855 xmlOutputBufferWriteString(buf
, (const char *)cur
->name
);
856 if (cur
->content
!= NULL
) {
857 xmlOutputBufferWriteString(buf
, " ");
858 xmlOutputBufferWriteString(buf
, (const char *)cur
->content
);
860 xmlOutputBufferWriteString(buf
, ">");
863 if (cur
->type
== HTML_ENTITY_REF_NODE
) {
864 xmlOutputBufferWriteString(buf
, "&");
865 xmlOutputBufferWriteString(buf
, (const char *)cur
->name
);
866 xmlOutputBufferWriteString(buf
, ";");
869 if (cur
->type
== HTML_PRESERVE_NODE
) {
870 if (cur
->content
!= NULL
) {
871 xmlOutputBufferWriteString(buf
, (const char *)cur
->content
);
877 * Get specific HTML info for that node.
880 info
= htmlTagLookup(cur
->name
);
884 xmlOutputBufferWriteString(buf
, "<");
885 if ((cur
->ns
!= NULL
) && (cur
->ns
->prefix
!= NULL
)) {
886 xmlOutputBufferWriteString(buf
, (const char *)cur
->ns
->prefix
);
887 xmlOutputBufferWriteString(buf
, ":");
889 xmlOutputBufferWriteString(buf
, (const char *)cur
->name
);
891 xmlNsListDumpOutput(buf
, cur
->nsDef
);
892 if (cur
->properties
!= NULL
)
893 htmlAttrListDumpOutput(buf
, doc
, cur
->properties
, encoding
);
895 if ((info
!= NULL
) && (info
->empty
)) {
896 xmlOutputBufferWriteString(buf
, ">");
897 if ((format
) && (!info
->isinline
) && (cur
->next
!= NULL
)) {
898 if ((cur
->next
->type
!= HTML_TEXT_NODE
) &&
899 (cur
->next
->type
!= HTML_ENTITY_REF_NODE
) &&
900 (cur
->parent
!= NULL
) &&
901 (cur
->parent
->name
!= NULL
) &&
902 (cur
->parent
->name
[0] != 'p')) /* p, pre, param */
903 xmlOutputBufferWriteString(buf
, "\n");
907 if (((cur
->type
== XML_ELEMENT_NODE
) || (cur
->content
== NULL
)) &&
908 (cur
->children
== NULL
)) {
909 if ((info
!= NULL
) && (info
->saveEndTag
!= 0) &&
910 (xmlStrcmp(BAD_CAST info
->name
, BAD_CAST
"html")) &&
911 (xmlStrcmp(BAD_CAST info
->name
, BAD_CAST
"body"))) {
912 xmlOutputBufferWriteString(buf
, ">");
914 xmlOutputBufferWriteString(buf
, "></");
915 if ((cur
->ns
!= NULL
) && (cur
->ns
->prefix
!= NULL
)) {
916 xmlOutputBufferWriteString(buf
, (const char *)cur
->ns
->prefix
);
917 xmlOutputBufferWriteString(buf
, ":");
919 xmlOutputBufferWriteString(buf
, (const char *)cur
->name
);
920 xmlOutputBufferWriteString(buf
, ">");
922 if ((format
) && (cur
->next
!= NULL
) &&
923 (info
!= NULL
) && (!info
->isinline
)) {
924 if ((cur
->next
->type
!= HTML_TEXT_NODE
) &&
925 (cur
->next
->type
!= HTML_ENTITY_REF_NODE
) &&
926 (cur
->parent
!= NULL
) &&
927 (cur
->parent
->name
!= NULL
) &&
928 (cur
->parent
->name
[0] != 'p')) /* p, pre, param */
929 xmlOutputBufferWriteString(buf
, "\n");
933 xmlOutputBufferWriteString(buf
, ">");
934 if ((cur
->type
!= XML_ELEMENT_NODE
) &&
935 (cur
->content
!= NULL
)) {
937 * Uses the OutputBuffer property to automatically convert
938 * invalids to charrefs
941 xmlOutputBufferWriteString(buf
, (const char *) cur
->content
);
943 if (cur
->children
!= NULL
) {
944 if ((format
) && (info
!= NULL
) && (!info
->isinline
) &&
945 (cur
->children
->type
!= HTML_TEXT_NODE
) &&
946 (cur
->children
->type
!= HTML_ENTITY_REF_NODE
) &&
947 (cur
->children
!= cur
->last
) &&
948 (cur
->name
!= NULL
) &&
949 (cur
->name
[0] != 'p')) /* p, pre, param */
950 xmlOutputBufferWriteString(buf
, "\n");
951 htmlNodeListDumpOutput(buf
, doc
, cur
->children
, encoding
, format
);
952 if ((format
) && (info
!= NULL
) && (!info
->isinline
) &&
953 (cur
->last
->type
!= HTML_TEXT_NODE
) &&
954 (cur
->last
->type
!= HTML_ENTITY_REF_NODE
) &&
955 (cur
->children
!= cur
->last
) &&
956 (cur
->name
!= NULL
) &&
957 (cur
->name
[0] != 'p')) /* p, pre, param */
958 xmlOutputBufferWriteString(buf
, "\n");
960 xmlOutputBufferWriteString(buf
, "</");
961 if ((cur
->ns
!= NULL
) && (cur
->ns
->prefix
!= NULL
)) {
962 xmlOutputBufferWriteString(buf
, (const char *)cur
->ns
->prefix
);
963 xmlOutputBufferWriteString(buf
, ":");
965 xmlOutputBufferWriteString(buf
, (const char *)cur
->name
);
966 xmlOutputBufferWriteString(buf
, ">");
967 if ((format
) && (info
!= NULL
) && (!info
->isinline
) &&
968 (cur
->next
!= NULL
)) {
969 if ((cur
->next
->type
!= HTML_TEXT_NODE
) &&
970 (cur
->next
->type
!= HTML_ENTITY_REF_NODE
) &&
971 (cur
->parent
!= NULL
) &&
972 (cur
->parent
->name
!= NULL
) &&
973 (cur
->parent
->name
[0] != 'p')) /* p, pre, param */
974 xmlOutputBufferWriteString(buf
, "\n");
979 * htmlNodeDumpOutput:
980 * @buf: the HTML buffer output
982 * @cur: the current node
983 * @encoding: the encoding string
985 * Dump an HTML node, recursive behaviour,children are printed too,
986 * and formatting returns/spaces are added.
989 htmlNodeDumpOutput(xmlOutputBufferPtr buf
, xmlDocPtr doc
,
990 xmlNodePtr cur
, const char *encoding
) {
991 htmlNodeDumpFormatOutput(buf
, doc
, cur
, encoding
, 1);
995 * htmlDocContentDumpFormatOutput:
996 * @buf: the HTML buffer output
998 * @encoding: the encoding string
999 * @format: should formatting spaces been added
1001 * Dump an HTML document.
1004 htmlDocContentDumpFormatOutput(xmlOutputBufferPtr buf
, xmlDocPtr cur
,
1005 const char *encoding
, int format
) {
1010 if ((buf
== NULL
) || (cur
== NULL
))
1014 * force to output the stuff as HTML, especially for entities
1017 cur
->type
= XML_HTML_DOCUMENT_NODE
;
1018 if (cur
->intSubset
!= NULL
) {
1019 htmlDtdDumpOutput(buf
, cur
, NULL
);
1021 if (cur
->children
!= NULL
) {
1022 htmlNodeListDumpOutput(buf
, cur
, cur
->children
, encoding
, format
);
1024 xmlOutputBufferWriteString(buf
, "\n");
1025 cur
->type
= (xmlElementType
) type
;
1029 * htmlDocContentDumpOutput:
1030 * @buf: the HTML buffer output
1031 * @cur: the document
1032 * @encoding: the encoding string
1034 * Dump an HTML document. Formating return/spaces are added.
1037 htmlDocContentDumpOutput(xmlOutputBufferPtr buf
, xmlDocPtr cur
,
1038 const char *encoding
) {
1039 htmlDocContentDumpFormatOutput(buf
, cur
, encoding
, 1);
1042 /************************************************************************
1044 * Saving functions front-ends *
1046 ************************************************************************/
1051 * @cur: the document
1053 * Dump an HTML document to an open FILE.
1055 * returns: the number of byte written or -1 in case of failure.
1058 htmlDocDump(FILE *f
, xmlDocPtr cur
) {
1059 xmlOutputBufferPtr buf
;
1060 xmlCharEncodingHandlerPtr handler
= NULL
;
1061 const char *encoding
;
1066 if ((cur
== NULL
) || (f
== NULL
)) {
1070 encoding
= (const char *) htmlGetMetaEncoding(cur
);
1072 if (encoding
!= NULL
) {
1073 xmlCharEncoding enc
;
1075 enc
= xmlParseCharEncoding(encoding
);
1076 if (enc
!= cur
->charset
) {
1077 if (cur
->charset
!= XML_CHAR_ENCODING_UTF8
) {
1084 handler
= xmlFindCharEncodingHandler(encoding
);
1085 if (handler
== NULL
)
1086 htmlSaveErr(XML_SAVE_UNKNOWN_ENCODING
, NULL
, encoding
);
1088 handler
= xmlFindCharEncodingHandler(encoding
);
1093 * Fallback to HTML or ASCII when the encoding is unspecified
1095 if (handler
== NULL
)
1096 handler
= xmlFindCharEncodingHandler("HTML");
1097 if (handler
== NULL
)
1098 handler
= xmlFindCharEncodingHandler("ascii");
1100 buf
= xmlOutputBufferCreateFile(f
, handler
);
1101 if (buf
== NULL
) return(-1);
1102 htmlDocContentDumpOutput(buf
, cur
, NULL
);
1104 ret
= xmlOutputBufferClose(buf
);
1110 * @filename: the filename (or URL)
1111 * @cur: the document
1113 * Dump an HTML document to a file. If @filename is "-" the stdout file is
1115 * returns: the number of byte written or -1 in case of failure.
1118 htmlSaveFile(const char *filename
, xmlDocPtr cur
) {
1119 xmlOutputBufferPtr buf
;
1120 xmlCharEncodingHandlerPtr handler
= NULL
;
1121 const char *encoding
;
1124 if ((cur
== NULL
) || (filename
== NULL
))
1129 encoding
= (const char *) htmlGetMetaEncoding(cur
);
1131 if (encoding
!= NULL
) {
1132 xmlCharEncoding enc
;
1134 enc
= xmlParseCharEncoding(encoding
);
1135 if (enc
!= cur
->charset
) {
1136 if (cur
->charset
!= XML_CHAR_ENCODING_UTF8
) {
1143 handler
= xmlFindCharEncodingHandler(encoding
);
1144 if (handler
== NULL
)
1145 htmlSaveErr(XML_SAVE_UNKNOWN_ENCODING
, NULL
, encoding
);
1150 * Fallback to HTML or ASCII when the encoding is unspecified
1152 if (handler
== NULL
)
1153 handler
= xmlFindCharEncodingHandler("HTML");
1154 if (handler
== NULL
)
1155 handler
= xmlFindCharEncodingHandler("ascii");
1158 * save the content to a temp buffer.
1160 buf
= xmlOutputBufferCreateFilename(filename
, handler
, cur
->compression
);
1161 if (buf
== NULL
) return(0);
1163 htmlDocContentDumpOutput(buf
, cur
, NULL
);
1165 ret
= xmlOutputBufferClose(buf
);
1170 * htmlSaveFileFormat:
1171 * @filename: the filename
1172 * @cur: the document
1173 * @format: should formatting spaces been added
1174 * @encoding: the document encoding
1176 * Dump an HTML document to a file using a given encoding.
1178 * returns: the number of byte written or -1 in case of failure.
1181 htmlSaveFileFormat(const char *filename
, xmlDocPtr cur
,
1182 const char *encoding
, int format
) {
1183 xmlOutputBufferPtr buf
;
1184 xmlCharEncodingHandlerPtr handler
= NULL
;
1187 if ((cur
== NULL
) || (filename
== NULL
))
1192 if (encoding
!= NULL
) {
1193 xmlCharEncoding enc
;
1195 enc
= xmlParseCharEncoding(encoding
);
1196 if (enc
!= cur
->charset
) {
1197 if (cur
->charset
!= XML_CHAR_ENCODING_UTF8
) {
1204 handler
= xmlFindCharEncodingHandler(encoding
);
1205 if (handler
== NULL
)
1206 htmlSaveErr(XML_SAVE_UNKNOWN_ENCODING
, NULL
, encoding
);
1208 htmlSetMetaEncoding(cur
, (const xmlChar
*) encoding
);
1210 htmlSetMetaEncoding(cur
, (const xmlChar
*) "UTF-8");
1214 * Fallback to HTML or ASCII when the encoding is unspecified
1216 if (handler
== NULL
)
1217 handler
= xmlFindCharEncodingHandler("HTML");
1218 if (handler
== NULL
)
1219 handler
= xmlFindCharEncodingHandler("ascii");
1222 * save the content to a temp buffer.
1224 buf
= xmlOutputBufferCreateFilename(filename
, handler
, 0);
1225 if (buf
== NULL
) return(0);
1227 htmlDocContentDumpFormatOutput(buf
, cur
, encoding
, format
);
1229 ret
= xmlOutputBufferClose(buf
);
1235 * @filename: the filename
1236 * @cur: the document
1237 * @encoding: the document encoding
1239 * Dump an HTML document to a file using a given encoding
1240 * and formatting returns/spaces are added.
1242 * returns: the number of byte written or -1 in case of failure.
1245 htmlSaveFileEnc(const char *filename
, xmlDocPtr cur
, const char *encoding
) {
1246 return(htmlSaveFileFormat(filename
, cur
, encoding
, 1));
1249 #endif /* LIBXML_OUTPUT_ENABLED */
1251 #define bottom_HTMLtree
1252 #include "elfgcchack.h"
1253 #endif /* LIBXML_HTML_ENABLED */