work done on auto-opening of <p> tags and cleanup of SAX output, Daniel.

2025-03-06 20:58:17 +03:00 · 2000-08-19 19:52:36 +00:00 · 2000-08-19 19:52:36 +00:00 · b8f25c9118
commit b8f25c9118
parent 979e55e889
23 changed files with 373 additions and 40 deletions
--- a/5
+++ b/5
@ -1,3 +1,8 @@
+Sat Aug 19 21:02:08 CEST 2000 Daniel Veillard <Daniel.Veillard@w3.org>
+
+	* HTMLparser.c SAX.c tree.c HTMLtree.h result/HTML/*: work
+	  done on auto-opening of <p> tags and cleanup of SAX output
+
 Sat Aug 19 18:45:40 CEST 2000 Daniel Veillard <Daniel.Veillard@w3.org>

 	* libxml.4  xmllint.1 Makefile.am libxml.spec.in: added man pages
--- a/HTMLparser.c
+++ b/HTMLparser.c
@ -552,6 +552,20 @@ char *htmlStartClose[] = {
 NULL
 };

+/*
+ * The list of HTML elements which are supposed not to have
+ * CDATA content and where a p element will be implied
+ *
+ * TODO: extend that list by reading the HTML SGML DtD on
+ *       implied paragraph
+ */
+static char *htmlNoContentElements[] = {
+    "html",
+    "head",
+    "body",
+    NULL
+};
+

 static char** htmlStartCloseIndex[100];
 static int htmlStartCloseIndexinitialized = 0;
@ -845,6 +859,49 @@ htmlCheckImplied(htmlParserCtxtPtr ctxt, const xmlChar *newtag) {
    }
 }

+/**
+ * htmlCheckParagraph
+ * @ctxt:  an HTML parser context
+ *
+ * Check whether a p element need to be implied before inserting
+ * characters in the current element.
+ *
+ * Returns 1 if a paragraph has been inserted, 0 if not and -1
+ *         in case of error.
+ */
+
+int
+htmlCheckParagraph(htmlParserCtxtPtr ctxt) {
+    const xmlChar *tag;
+    int i;
+
+    if (ctxt == NULL)
+	return(-1);
+    tag = ctxt->name;
+    if (tag == NULL) {
+	htmlAutoClose(ctxt, BAD_CAST"p");
+	htmlCheckImplied(ctxt, BAD_CAST"p");
+	htmlnamePush(ctxt, xmlStrdup(BAD_CAST"p"));
+	if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
+	    ctxt->sax->startElement(ctxt->userData, BAD_CAST"p", NULL);
+	return(1);
+    }
+    for (i = 0; htmlNoContentElements[i] != NULL; i++) {
+	if (!xmlStrcmp(tag, BAD_CAST htmlNoContentElements[i])) {
+#ifdef DEBUG
+	    fprintf(stderr,"Implied element paragraph\n");
+#endif    
+	    htmlAutoClose(ctxt, BAD_CAST"p");
+	    htmlCheckImplied(ctxt, BAD_CAST"p");
+	    htmlnamePush(ctxt, xmlStrdup(BAD_CAST"p"));
+	    if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
+		ctxt->sax->startElement(ctxt->userData, BAD_CAST"p", NULL);
+	    return(1);
+	}
+    }
+    return(0);
+}
+
 /************************************************************************
 *									*
 * 		The list of HTML predefined entities			*
@ -1253,7 +1310,8 @@ UTF8ToHtml(unsigned char* out, int *outlen,
 			    sizeof(html40EntitiesTable[0]));i++) {
 		if (html40EntitiesTable[i].value == c) {
 #ifdef DEBUG
-		    fprintf(stderr,"Found entity %s\n", name);
+		    fprintf(stderr,"Found entity %s\n", 
+			    html40EntitiesTable[i].name);
 #endif
 		    goto found_ent;
 		}
@ -1496,20 +1554,21 @@ htmlHandleEntity(htmlParserCtxtPtr ctxt, xmlEntityPtr entity) {
    /*
     * Just handle the content as a set of chars.
     */
+    htmlCheckParagraph(ctxt);
    if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
 	ctxt->sax->characters(ctxt->userData, entity->content, len);

 }

 /**
- * htmlNewDoc:
+ * htmlNewDocNoDtD:
 * @URI:  URI for the dtd, or NULL
 * @ExternalID:  the external ID of the DTD, or NULL
 *
- * Returns a new document
+ * Returns a new document, do not intialize the DTD if not provided
 */
 htmlDocPtr
-htmlNewDoc(const xmlChar *URI, const xmlChar *ExternalID) {
+htmlNewDocNoDtD(const xmlChar *URI, const xmlChar *ExternalID) {
    xmlDocPtr cur;

    /*
@ -1525,12 +1584,8 @@ htmlNewDoc(const xmlChar *URI, const xmlChar *ExternalID) {
    cur->type = XML_HTML_DOCUMENT_NODE;
    cur->version = NULL;
    cur->intSubset = NULL;
-    if ((ExternalID == NULL) &&
-	(URI == NULL))
-	xmlCreateIntSubset(cur, BAD_CAST "HTML",
-		    BAD_CAST "-//W3C//DTD HTML 4.0 Transitional//EN",
-		    BAD_CAST "http://www.w3.org/TR/REC-html40/loose.dtd");
-    else
+    if ((ExternalID != NULL) ||
+	(URI != NULL))
 	xmlCreateIntSubset(cur, BAD_CAST "HTML", ExternalID, URI);
    cur->doc = cur;
    cur->name = NULL;
@ -1548,6 +1603,23 @@ htmlNewDoc(const xmlChar *URI, const xmlChar *ExternalID) {
    return(cur);
 }

+/**
+ * htmlNewDoc:
+ * @URI:  URI for the dtd, or NULL
+ * @ExternalID:  the external ID of the DTD, or NULL
+ *
+ * Returns a new document
+ */
+htmlDocPtr
+htmlNewDoc(const xmlChar *URI, const xmlChar *ExternalID) {
+    if ((URI == NULL) && (ExternalID == NULL))
+	return(htmlNewDocNoDtD(
+		    BAD_CAST "-//W3C//DTD HTML 4.0 Transitional//EN",
+		    BAD_CAST "http://www.w3.org/TR/REC-html40/loose.dtd"));
+
+    return(htmlNewDocNoDtD(URI, ExternalID));
+}
+

 /************************************************************************
 *									*
@ -2062,6 +2134,7 @@ htmlParseCharData(htmlParserCtxtPtr ctxt, int cdata) {
 			ctxt->sax->ignorableWhitespace(ctxt->userData,
 			                               buf, nbchar);
 		} else {
+		    htmlCheckParagraph(ctxt);
 		    if (ctxt->sax->characters != NULL)
 			ctxt->sax->characters(ctxt->userData, buf, nbchar);
 		}
@ -2080,6 +2153,7 @@ htmlParseCharData(htmlParserCtxtPtr ctxt, int cdata) {
 		if (ctxt->sax->ignorableWhitespace != NULL)
 		    ctxt->sax->ignorableWhitespace(ctxt->userData, buf, nbchar);
 	    } else {
+		htmlCheckParagraph(ctxt);
 		if (ctxt->sax->characters != NULL)
 		    ctxt->sax->characters(ctxt->userData, buf, nbchar);
 	    }
@ -2861,16 +2935,19 @@ htmlParseReference(htmlParserCtxtPtr ctxt) {
        }
 	out[i] = 0;

+	htmlCheckParagraph(ctxt);
 	if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
 	    ctxt->sax->characters(ctxt->userData, out, i);
    } else {
 	ent = htmlParseEntityRef(ctxt, &name);
 	if (name == NULL) {
+	    htmlCheckParagraph(ctxt);
 	    if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
 	        ctxt->sax->characters(ctxt->userData, BAD_CAST "&", 1);
 	    return;
 	}
 	if ((ent == NULL) || (ent->value <= 0)) {
+	    htmlCheckParagraph(ctxt);
 	    if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL)) {
 		ctxt->sax->characters(ctxt->userData, BAD_CAST "&", 1);
 		ctxt->sax->characters(ctxt->userData, name, xmlStrlen(name));
@ -2895,6 +2972,7 @@ htmlParseReference(htmlParserCtxtPtr ctxt) {
 	    }
 	    out[i] = 0;

+	    htmlCheckParagraph(ctxt);
 	    if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
 		ctxt->sax->characters(ctxt->userData, out, i);
 	}
@ -2941,6 +3019,21 @@ htmlParseContent(htmlParserCtxtPtr ctxt) {
 	    return;
 	}

+	/*
+	 * Sometimes DOCTYPE arrives in the middle of the document
+	 */
+	if ((CUR == '<') && (NXT(1) == '!') &&
+	    (UPP(2) == 'D') && (UPP(3) == 'O') &&
+	    (UPP(4) == 'C') && (UPP(5) == 'T') &&
+	    (UPP(6) == 'Y') && (UPP(7) == 'P') &&
+	    (UPP(8) == 'E')) {
+	    if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+		ctxt->sax->error(ctxt->userData,
+		     "Misplaced DOCTYPE declaration\n");
+	    ctxt->wellFormed = 0;
+	    htmlParseDocTypeDecl(ctxt);
+	}
+
 	/*
 	 * First case :  a comment
 	 */
@ -3185,6 +3278,8 @@ htmlParseElement(htmlParserCtxtPtr ctxt) {

 int
 htmlParseDocument(htmlParserCtxtPtr ctxt) {
+    xmlDtdPtr dtd;
+
    htmlDefaultSAXHandlerInit();
    ctxt->html = 1;

@ -3258,6 +3353,15 @@ htmlParseDocument(htmlParserCtxtPtr ctxt) {
     */
    if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
        ctxt->sax->endDocument(ctxt->userData);
+
+    if (ctxt->myDoc != NULL) {
+	dtd = xmlGetIntSubset(ctxt->myDoc);
+	if (dtd == NULL)
+	    ctxt->myDoc->intSubset = 
+		xmlCreateIntSubset(ctxt->myDoc, BAD_CAST "HTML", 
+		    BAD_CAST "-//W3C//DTD HTML 4.0 Transitional//EN",
+		    BAD_CAST "http://www.w3.org/TR/REC-html40/loose.dtd");
+    }
    if (! ctxt->wellFormed) return(-1);
    return(0);
 }
@ -3848,6 +3952,7 @@ htmlParseTryOrFinish(htmlParserCtxtPtr ctxt, int terminate) {
 		    xmlChar chr[2] = { 0 , 0 } ;

 		    chr[0] = (xmlChar) ctxt->token;
+		    htmlCheckParagraph(ctxt);
 		    if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
 			ctxt->sax->characters(ctxt->userData, chr, 1);
 		    ctxt->token = 0;
@ -3862,6 +3967,7 @@ htmlParseTryOrFinish(htmlParserCtxtPtr ctxt, int terminate) {
 				    ctxt->sax->ignorableWhitespace(
 					    ctxt->userData, &cur, 1);
 			    } else {
+				htmlCheckParagraph(ctxt);
 				if (ctxt->sax->characters != NULL)
 				    ctxt->sax->characters(
 					    ctxt->userData, &cur, 1);
@ -3878,7 +3984,23 @@ htmlParseTryOrFinish(htmlParserCtxtPtr ctxt, int terminate) {
 		cur = in->cur[0];
 		next = in->cur[1];
 		cons = ctxt->nbChars;
-	        if ((cur == '<') && (next == '!') &&
+		/*
+		 * Sometimes DOCTYPE arrives in the middle of the document
+		 */
+		if ((cur == '<') && (next == '!') &&
+		    (UPP(2) == 'D') && (UPP(3) == 'O') &&
+		    (UPP(4) == 'C') && (UPP(5) == 'T') &&
+		    (UPP(6) == 'Y') && (UPP(7) == 'P') &&
+		    (UPP(8) == 'E')) {
+		    if ((!terminate) &&
+		        (htmlParseLookupSequence(ctxt, '>', 0, 0) < 0))
+			goto done;
+		    if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+			ctxt->sax->error(ctxt->userData,
+			     "Misplaced DOCTYPE declaration\n");
+		    ctxt->wellFormed = 0;
+		    htmlParseDocTypeDecl(ctxt);
+		} else if ((cur == '<') && (next == '!') &&
 		    (in->cur[2] == '-') && (in->cur[3] == '-')) {
 		    if ((!terminate) &&
 		        (htmlParseLookupSequence(ctxt, '-', '-', '>') < 0))
@ -4040,6 +4162,17 @@ done:
 		ctxt->sax->endDocument(ctxt->userData);
 	}
    }
+    if ((ctxt->myDoc != NULL) &&
+	((terminate) || (ctxt->instate == XML_PARSER_EOF) ||
+	 (ctxt->instate == XML_PARSER_EPILOG))) {
+	xmlDtdPtr dtd;
+	dtd = xmlGetIntSubset(ctxt->myDoc);
+	if (dtd == NULL)
+	    ctxt->myDoc->intSubset = 
+		xmlCreateIntSubset(ctxt->myDoc, BAD_CAST "HTML", 
+		    BAD_CAST "-//W3C//DTD HTML 4.0 Transitional//EN",
+		    BAD_CAST "http://www.w3.org/TR/REC-html40/loose.dtd");
+    }
 #ifdef DEBUG_PUSH
    fprintf(stderr, "HPP: done %d\n", ret);
 #endif
--- a/HTMLtree.h
+++ b/HTMLtree.h
@ -25,6 +25,8 @@ extern "C" {

 htmlDocPtr	htmlNewDoc		(const xmlChar *URI,
 					 const xmlChar *ExternalID);
+htmlDocPtr	htmlNewDocNoDtD		(const xmlChar *URI,
+					 const xmlChar *ExternalID);
 const xmlChar *	htmlGetMetaEncoding	(htmlDocPtr doc);
 int		htmlSetMetaEncoding	(htmlDocPtr doc,
 					 const xmlChar *encoding);
--- a/SAX.c
+++ b/SAX.c
@ -169,6 +169,8 @@ internalSubset(void *ctx, const xmlChar *name,
 	return;
    dtd = xmlGetIntSubset(ctxt->myDoc);
    if (dtd != NULL) {
+	if (ctxt->html)
+	    return;
 	xmlUnlinkNode((xmlNodePtr) dtd);
 	xmlFreeDtd(dtd);
 	ctxt->myDoc->intSubset = NULL;
@ -605,7 +607,7 @@ startDocument(void *ctx)
    if (ctxt->html) {
 	if (ctxt->myDoc == NULL)
 #ifdef LIBXML_HTML_ENABLED
-	    ctxt->myDoc = htmlNewDoc(NULL, NULL);
+	    ctxt->myDoc = htmlNewDocNoDtD(NULL, NULL);
 #else
        fprintf(stderr, "libxml2 built without HTML support\n");
 #endif
--- a/include/libxml/HTMLtree.h
+++ b/include/libxml/HTMLtree.h
@ -25,6 +25,8 @@ extern "C" {

 htmlDocPtr	htmlNewDoc		(const xmlChar *URI,
 					 const xmlChar *ExternalID);
+htmlDocPtr	htmlNewDocNoDtD		(const xmlChar *URI,
+					 const xmlChar *ExternalID);
 const xmlChar *	htmlGetMetaEncoding	(htmlDocPtr doc);
 int		htmlSetMetaEncoding	(htmlDocPtr doc,
 					 const xmlChar *encoding);
--- a/result/HTML/Down.html
+++ b/result/HTML/Down.html
@ -3,8 +3,10 @@
 <head><title>This service is temporary down</title></head>
 <body bgcolor="#FFFFFF">
 <h1 align="center">Sorry, this service is temporary down</h1>
+<p>
 We are doing our best to get it back on-line,

+</p>
 <p>The W3C system administrators</p>
 </body>
 </html>
--- a/result/HTML/Down.html.sax
+++ b/result/HTML/Down.html.sax
@ -1,36 +1,53 @@
 SAX.setDocumentLocator()
 SAX.startDocument()
 SAX.startElement(html)
+SAX.startElement(body)
+SAX.startElement(p)
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.startElement(head)
+SAX.endElement(head)
+SAX.startElement(p)
 SAX.characters(
  , 3)
+SAX.endElement(p)
 SAX.startElement(title)
 SAX.characters(This service is temporary down, 30)
 SAX.endElement(title)
+SAX.startElement(p)
 SAX.characters(
 , 1)
-SAX.endElement(head)
+SAX.error: Unexpected end tag : head
 SAX.characters(

 , 2)
+SAX.endElement(p)
 SAX.startElement(body, bgcolor='#FFFFFF')
+SAX.startElement(p)
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.startElement(h1, align='center')
 SAX.characters(Sorry, this service is tempora, 37)
 SAX.endElement(h1)
+SAX.startElement(p)
 SAX.characters(
 We are doing our best to get , 48)
+SAX.endElement(p)
 SAX.startElement(p)
 SAX.characters(The W3C system administrators, 29)
 SAX.endElement(p)
+SAX.startElement(p)
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.endElement(body)
+SAX.startElement(p)
 SAX.characters(
 , 1)
+SAX.endElement(p)
+SAX.endElement(body)
 SAX.endElement(html)
 SAX.ignorableWhitespace(
 , 1)
--- a/result/HTML/doc2.htm
+++ b/result/HTML/doc2.htm
@ -8,7 +8,7 @@
          function NS_NewOpen(url,nam,atr){return(new NS_NullWindow());}
          window.open=NS_NewOpen;
 </script>
-<!-- END Naviscope Javascript -->!DOCTYPE HTML PUBLIC &quot;-//W3C//DTD HTML 4.0 Transitional//EN&quot;&gt;<!-- saved from url=(0027)http://www.agents-tech.com/ --><meta content="text/html; charset=iso-8859-1" http-equiv="Content-Type">
+<!-- END Naviscope Javascript --><!-- saved from url=(0027)http://www.agents-tech.com/ --><meta content="text/html; charset=iso-8859-1" http-equiv="Content-Type">
 <meta content="Copernic.com Inc. develops innovative agent technology solutions to efficiently access and manage the overwhelming quantity of information available on the Internet and intranets." name="DESCRIPTION">
 <meta content="agent,technology,intranet,extranet,management,filtering,ranking,solution,service,intelligent,intelligence,client,server,architecture,developer,development,information,telecommunication,announcement,press,product,profile,contact,multi-agent,meta-search,metasearch,multi-thread,mobile,wireless,shopping,robot,PCS,Copernic,engine,toolkit,CDK,EDK" name="KEYWORDS">
 <meta content="MSHTML 5.00.3103.1000" name="GENERATOR">
--- a/result/HTML/doc2.htm.err
+++ b/result/HTML/doc2.htm.err
@ -1,3 +1,3 @@
-./test/HTML/doc2.htm:10: error: htmlParseStartTag: invalid element name
+./test/HTML/doc2.htm:10: error: Misplaced DOCTYPE declaration
 <!-- END Naviscope Javascript --><!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Tr
-                                  ^
+                                 ^
--- a/result/HTML/doc2.htm.sax
+++ b/result/HTML/doc2.htm.sax
@ -8,6 +8,9 @@ SAX.startElement(head)
 SAX.startElement(title)
 SAX.characters(Welcome to Copernic.com, 23)
 SAX.endElement(title)
+SAX.endElement(head)
+SAX.startElement(body)
+SAX.startElement(p)
 SAX.characters(
 , 1)
 SAX.startElement(script, language='javascript')
@ -17,8 +20,8 @@ SAX.endElement(script)
 SAX.characters(
 , 1)
 SAX.comment( END Naviscope Javascript )
-SAX.error: htmlParseStartTag: invalid element name
-SAX.characters(!DOCTYPE HTML PUBLIC "-//W3C//, 61)
+SAX.error: Misplaced DOCTYPE declaration
+SAX.internalSubset(HTML, -//W3C//DTD HTML 4.0 Transitional//EN, )
 SAX.comment( saved from url=(0027)http://www.agents-tech.com/ )
 SAX.characters(
 , 1)
@ -36,8 +39,7 @@ SAX.characters(
 , 1)
 SAX.startElement(meta, content='MSHTML 5.00.3103.1000', name='GENERATOR')
 SAX.endElement(meta)
-SAX.endElement(head)
-SAX.startElement(body)
+SAX.error: Unexpected end tag : head
 SAX.startElement(frameset, border='false', cols='172,*', frameBorder='0', frameSpacing='0')
 SAX.startElement(frame, marginHeight='0', marginWidth='0', name='left', noResize, scrolling='no', src='doc2_files/side.htm', target='rtop')
 SAX.endElement(frame)
@ -52,18 +54,23 @@ SAX.characters(

  , 4)
 SAX.startElement(body, bgcolor='#FFFFFF', text='#000000', link='#000080', vlink='#000080', alink='#000080', topmargin='0', leftmargin='0', marginheight='0', marginwidth='0')
+SAX.startElement(p)
 SAX.characters(
  , 3)
+SAX.endElement(p)
 SAX.startElement(p)
 SAX.characters(This page uses frames, but you, 61)
 SAX.endElement(p)
+SAX.startElement(p)
 SAX.characters(
  , 3)
+SAX.endElement(p)
 SAX.endElement(body)
 SAX.characters(
  , 3)
 SAX.endElement(noframes)
 SAX.endElement(frameset)
+SAX.endElement(p)
 SAX.endElement(body)
 SAX.endElement(html)
 SAX.ignorableWhitespace(
--- a/result/HTML/doc3.htm
+++ b/result/HTML/doc3.htm
@ -8,7 +8,7 @@
          function NS_NewOpen(url,nam,atr){return(new NS_NullWindow());}
          window.open=NS_NewOpen;
 </script>
-<!-- END Naviscope Javascript -->!DOCTYPE HTML PUBLIC &quot;-//W3C//DTD HTML 3.2//EN&quot;&gt;<!--last modified on Tuesday, February 22, 2000 11:47 PM --><meta content="text/html;CHARSET=iso-8859-1" http-equiv="Content-Type">
+<!-- END Naviscope Javascript --><!--last modified on Tuesday, February 22, 2000 11:47 PM --><meta content="text/html;CHARSET=iso-8859-1" http-equiv="Content-Type">
 <meta content="Tim" name="Author">
 <style type="text/css">A.nav {
 	COLOR: #003399; TEXT-DECORATION: none
--- a/result/HTML/doc3.htm.err
+++ b/result/HTML/doc3.htm.err
@ -1,6 +1,6 @@
-./test/HTML/doc3.htm:10: error: htmlParseStartTag: invalid element name
+./test/HTML/doc3.htm:10: error: Misplaced DOCTYPE declaration
 <!-- END Naviscope Javascript --><!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//E
-                                  ^
+                                 ^
 ./test/HTML/doc3.htm:52: error: htmlParseEntityRef: expecting ';'
 href="http://ads.gamesquad.net/addclick.exe/adclick.cgi?REGION=game|tech|ent&i
                                                                                ^
--- a/result/HTML/doc3.htm.sax
+++ b/result/HTML/doc3.htm.sax
@ -8,6 +8,9 @@ SAX.startElement(head)
 SAX.startElement(title)
 SAX.characters(BP6.com #1 online resource for, 47)
 SAX.endElement(title)
+SAX.endElement(head)
+SAX.startElement(body)
+SAX.startElement(p)
 SAX.characters(
 , 2)
 SAX.startElement(script, language='javascript')
@ -17,8 +20,8 @@ SAX.endElement(script)
 SAX.characters(
 , 2)
 SAX.comment( END Naviscope Javascript )
-SAX.error: htmlParseStartTag: invalid element name
-SAX.characters(!DOCTYPE HTML PUBLIC "-//W3C//, 48)
+SAX.error: Misplaced DOCTYPE declaration
+SAX.internalSubset(HTML, -//W3C//DTD HTML 3.2//EN, )
 SAX.comment(last modified on Tuesday, February 22, 2000 11:47 PM )
 SAX.characters(
 , 2)
@ -69,12 +72,15 @@ SAX.characters(
 , 4)
 SAX.startElement(meta, content='MSHTML 5.00.3103.1000', name='GENERATOR')
 SAX.endElement(meta)
-SAX.endElement(head)
+SAX.error: Unexpected end tag : head
 SAX.characters(
 , 2)
+SAX.endElement(p)
 SAX.startElement(body, aLink='red', bgColor='black', link='red', text='white', vLink='red')
+SAX.startElement(p)
 SAX.characters(
 , 2)
+SAX.endElement(p)
 SAX.startElement(p)
 SAX.characters(
 , 2)
@ -2913,6 +2919,7 @@ SAX.comment(	<TR>		<TD WIDTH="780">			<P ALIGN="CENTER"><FONT SIZE="1" COLOR="#9
 SAX.endElement(tbody)
 SAX.endElement(table)
 SAX.endElement(div)
+SAX.startElement(p)
 SAX.characters(
 , 2)
 SAX.startElement(script)
@ -2920,6 +2927,8 @@ SAX.characters( window.open=NS_ActualOpen; , 28)
 SAX.endElement(script)
 SAX.characters(
 , 2)
+SAX.endElement(p)
+SAX.endElement(body)
 SAX.endElement(body)
 SAX.endElement(html)
 SAX.ignorableWhitespace(
--- a/result/HTML/fp40.htm.sax
+++ b/result/HTML/fp40.htm.sax
@ -2,30 +2,39 @@ SAX.setDocumentLocator()
 SAX.startDocument()
 SAX.internalSubset(html, -//IETF//DTD HTML//EN, )
 SAX.startElement(html)
+SAX.startElement(body)
+SAX.startElement(p)
 SAX.characters(

 , 2)
+SAX.endElement(p)
 SAX.startElement(head)
+SAX.endElement(head)
+SAX.startElement(p)
 SAX.characters(
 , 1)
 SAX.startElement(meta, name='GENERATOR', content='Microsoft FrontPage 4.0')
 SAX.endElement(meta)
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.startElement(title)
 SAX.characters(README - Microsoft FrontPage 2, 51)
 SAX.endElement(title)
+SAX.startElement(p)
 SAX.characters(
 , 1)
 SAX.startElement(meta, name='Microsoft Theme', content='none')
 SAX.endElement(meta)
 SAX.characters(
 , 1)
-SAX.endElement(head)
+SAX.error: Unexpected end tag : head
 SAX.characters(

 , 2)
+SAX.endElement(p)
 SAX.startElement(body)
+SAX.startElement(p)
 SAX.characters(
 , 1)
 SAX.startElement(font, face='Verdana')
@ -169,6 +178,7 @@ SAX.characters(
 SAX.endElement(font)
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.startElement(blockquote)
 SAX.characters(
  , 3)
@ -184,6 +194,7 @@ SAX.endElement(font)
 SAX.characters(
 , 1)
 SAX.endElement(blockquote)
+SAX.startElement(p)
 SAX.characters(
 , 1)
 SAX.startElement(font, face='Verdana')
@ -454,9 +465,13 @@ SAX.characters(
 SAX.endElement(font)
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.endElement(body)
+SAX.startElement(p)
 SAX.characters(
 , 1)
+SAX.endElement(p)
+SAX.endElement(body)
 SAX.endElement(html)
 SAX.ignorableWhitespace(
 , 1)
--- a/result/HTML/liclose.html.sax
+++ b/result/HTML/liclose.html.sax
@ -2,21 +2,31 @@ SAX.setDocumentLocator()
 SAX.startDocument()
 SAX.internalSubset(HTML, -//W3C//DTD HTML 4.0 Transitional//EN, http://www.w3.org/TR/REC-html40/loose.dtd)
 SAX.startElement(html)
+SAX.startElement(body)
+SAX.startElement(p)
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.startElement(head)
+SAX.endElement(head)
+SAX.startElement(p)
 SAX.characters(
  , 3)
+SAX.endElement(p)
 SAX.startElement(title)
 SAX.endElement(title)
+SAX.startElement(p)
 SAX.characters(
 , 1)
-SAX.endElement(head)
+SAX.error: Unexpected end tag : head
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.startElement(body)
+SAX.startElement(p)
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.startElement(ul)
 SAX.characters(
 , 1)
@ -28,9 +38,12 @@ SAX.startElement(li)
 SAX.characters(Second item, closes the first , 34)
 SAX.endElement(li)
 SAX.endElement(ul)
+SAX.startElement(p)
 SAX.characters(

 , 2)
+SAX.endElement(p)
+SAX.endElement(body)
 SAX.endElement(body)
 SAX.endElement(html)
 SAX.ignorableWhitespace(
--- a/result/HTML/reg1.html.sax
+++ b/result/HTML/reg1.html.sax
@ -1,35 +1,50 @@
 SAX.setDocumentLocator()
 SAX.startDocument()
 SAX.startElement(html)
+SAX.startElement(body)
+SAX.startElement(p)
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.startElement(head)
+SAX.endElement(head)
+SAX.startElement(p)
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.startElement(title)
 SAX.characters(Regression test 1, 17)
 SAX.endElement(title)
+SAX.startElement(p)
 SAX.characters(
 , 1)
-SAX.endElement(head)
+SAX.error: Unexpected end tag : head
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.startElement(body)
+SAX.startElement(p)
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.startElement(h1)
 SAX.characters(Regression test 1, 17)
 SAX.endElement(h1)
+SAX.startElement(p)
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.startElement(p)
 SAX.characters(
 Ok file no problem
 , 20)
 SAX.endElement(p)
 SAX.endElement(body)
+SAX.startElement(p)
 SAX.characters(
 , 1)
+SAX.endElement(p)
+SAX.endElement(body)
 SAX.endElement(html)
 SAX.ignorableWhitespace(
 , 1)
--- a/result/HTML/reg2.html.sax
+++ b/result/HTML/reg2.html.sax
@ -1,27 +1,39 @@
 SAX.setDocumentLocator()
 SAX.startDocument()
 SAX.startElement(html)
+SAX.startElement(body)
+SAX.startElement(p)
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.startElement(head)
+SAX.endElement(head)
+SAX.startElement(p)
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.startElement(title)
 SAX.characters(Regression test 2, 17)
 SAX.endElement(title)
+SAX.startElement(p)
 SAX.characters(
 , 1)
-SAX.endElement(head)
+SAX.error: Unexpected end tag : head
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.startElement(body)
+SAX.startElement(p)
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.startElement(h1)
 SAX.characters(Regression test 2, 17)
 SAX.endElement(h1)
+SAX.startElement(p)
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.startElement(p)
 SAX.characters(
 Autoclose of tag P
@ -33,8 +45,11 @@ Ok file no problem
 , 20)
 SAX.endElement(p)
 SAX.endElement(body)
+SAX.startElement(p)
 SAX.characters(
 , 1)
+SAX.endElement(p)
+SAX.endElement(body)
 SAX.endElement(html)
 SAX.ignorableWhitespace(
 , 1)
--- a/result/HTML/reg3.html.sax
+++ b/result/HTML/reg3.html.sax
@ -1,27 +1,39 @@
 SAX.setDocumentLocator()
 SAX.startDocument()
 SAX.startElement(html)
+SAX.startElement(body)
+SAX.startElement(p)
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.startElement(head)
+SAX.endElement(head)
+SAX.startElement(p)
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.startElement(title)
 SAX.characters(Regression test 3, 17)
 SAX.endElement(title)
+SAX.startElement(p)
 SAX.characters(
 , 1)
-SAX.endElement(head)
+SAX.error: Unexpected end tag : head
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.startElement(body)
+SAX.startElement(p)
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.startElement(h1)
 SAX.characters(Regression test 3, 17)
 SAX.endElement(h1)
+SAX.startElement(p)
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.startElement(p)
 SAX.characters(
 Autoclose of tag P
@ -29,16 +41,21 @@ Autoclose of tag P
 SAX.endElement(p)
 SAX.startElement(hr)
 SAX.endElement(hr)
+SAX.startElement(p)
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.startElement(p)
 SAX.characters(
 Ok file no problem
 , 20)
 SAX.endElement(p)
 SAX.endElement(body)
+SAX.startElement(p)
 SAX.characters(
 , 1)
+SAX.endElement(p)
+SAX.endElement(body)
 SAX.endElement(html)
 SAX.ignorableWhitespace(
 , 1)
--- a/result/HTML/reg4.html.sax
+++ b/result/HTML/reg4.html.sax
@ -1,27 +1,39 @@
 SAX.setDocumentLocator()
 SAX.startDocument()
 SAX.startElement(html)
+SAX.startElement(body)
+SAX.startElement(p)
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.startElement(head)
+SAX.endElement(head)
+SAX.startElement(p)
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.startElement(title)
 SAX.characters(Regression test 4, 17)
 SAX.endElement(title)
+SAX.startElement(p)
 SAX.characters(
 , 1)
-SAX.endElement(head)
+SAX.error: Unexpected end tag : head
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.startElement(body)
+SAX.startElement(p)
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.startElement(h1)
 SAX.characters(Regression test 4, 17)
 SAX.endElement(h1)
+SAX.startElement(p)
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.startElement(p)
 SAX.characters(
 Wrong close of tag P
@ -29,14 +41,20 @@ Wrong close of tag P
 SAX.endElement(p)
 SAX.startElement(hr)
 SAX.endElement(hr)
+SAX.startElement(p)
 SAX.characters(
 , 1)
-SAX.error: Unexpected end tag : p
+SAX.endElement(p)
+SAX.startElement(p)
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.endElement(body)
+SAX.startElement(p)
 SAX.characters(
 , 1)
+SAX.endElement(p)
+SAX.endElement(body)
 SAX.endElement(html)
 SAX.ignorableWhitespace(
 , 1)
--- a/result/HTML/test2.html.sax
+++ b/result/HTML/test2.html.sax
@ -3,17 +3,25 @@ SAX.startDocument()
 SAX.internalSubset(HTML, -//W3C//DTD HTML 4.0 Transitional//EN, http://www.w3.org/TR/REC-html40/loose.dtd)
 SAX.startElement(html)
 SAX.startElement(head)
+SAX.endElement(head)
+SAX.startElement(body)
+SAX.startElement(p)
 SAX.characters( , 1)
+SAX.endElement(p)
 SAX.startElement(title)
 SAX.characters(Linux Today, 11)
 SAX.endElement(title)
-SAX.endElement(head)
+SAX.error: Unexpected end tag : head
+SAX.startElement(p)
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.startElement(body, bgcolor='White', link='Blue', text='Black', VLINK='Black', ALINK='Red')
+SAX.startElement(p)
 SAX.characters(

 , 2)
+SAX.endElement(p)
 SAX.startElement(center)
 SAX.characters(
 , 1)
@ -130,15 +138,20 @@ SAX.endElement(font)
 SAX.characters(
 , 1)
 SAX.endElement(center)
+SAX.startElement(p)
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.startElement(p)
 SAX.characters(
 , 1)
 SAX.endElement(p)
 SAX.endElement(body)
+SAX.startElement(p)
 SAX.characters(
 , 1)
+SAX.endElement(p)
+SAX.endElement(body)
 SAX.endElement(html)
 SAX.ignorableWhitespace(
 , 1)
--- a/result/HTML/test3.html.sax
+++ b/result/HTML/test3.html.sax
@ -1,19 +1,23 @@
 SAX.setDocumentLocator()
 SAX.startDocument()
 SAX.startElement(html)
+SAX.startElement(body)
+SAX.startElement(p)
 SAX.characters(
 , 2)
+SAX.endElement(p)
 SAX.startElement(head)
+SAX.endElement(head)
+SAX.startElement(p)
 SAX.characters(
 , 2)
 SAX.startElement(base, target='contents')
 SAX.endElement(base)
 SAX.characters(
 , 2)
-SAX.endElement(head)
+SAX.error: Unexpected end tag : head
 SAX.characters(
 , 2)
-SAX.startElement(body)
 SAX.startElement(a, name='ProblemDomain.Package')
 SAX.startElement(h2)
 SAX.characters(Component Package diagram Prob, 39)
@ -21,13 +25,16 @@ SAX.endElement(h2)
 SAX.characters(
 , 2)
 SAX.endElement(a)
+SAX.endElement(p)
 SAX.startElement(p)
 SAX.endElement(p)
 SAX.startElement(hr)
 SAX.endElement(hr)
 SAX.error: Unexpected end tag : p
+SAX.startElement(p)
 SAX.characters(
 , 2)
+SAX.endElement(p)
 SAX.startElement(dl)
 SAX.characters(
 , 2)
@ -59,15 +66,19 @@ SAX.characters(Interface, thats stores and ma, 58)
 SAX.endElement(dd)
 SAX.endElement(dd)
 SAX.endElement(dl)
+SAX.startElement(p)
 SAX.characters(
 , 2)
+SAX.endElement(p)
 SAX.startElement(p)
 SAX.endElement(p)
 SAX.startElement(hr)
 SAX.endElement(hr)
 SAX.error: Unexpected end tag : p
+SAX.startElement(p)
 SAX.characters(
 , 2)
+SAX.endElement(p)
 SAX.startElement(dl)
 SAX.characters(

@ -163,9 +174,11 @@ SAX.endElement(dt)
 SAX.characters(
 , 2)
 SAX.endElement(dl)
+SAX.startElement(p)
 SAX.characters(

 , 4)
+SAX.endElement(p)
 SAX.startElement(h4)
 SAX.startElement(b)
 SAX.characters(Links, 5)
@ -173,8 +186,10 @@ SAX.error: Opening and ending tag mismatch: h4 and b
 SAX.endElement(b)
 SAX.endElement(h4)
 SAX.error: Unexpected end tag : b
+SAX.startElement(p)
 SAX.characters(
 , 2)
+SAX.endElement(p)
 SAX.startElement(ul)
 SAX.startElement(li)
 SAX.startElement(b)
@ -185,12 +200,16 @@ SAX.characters(HumanInterface, 14)
 SAX.endElement(a)
 SAX.endElement(li)
 SAX.endElement(ul)
+SAX.startElement(p)
 SAX.characters(
 , 2)
+SAX.endElement(p)
 SAX.startElement(dir)
 SAX.endElement(dir)
+SAX.startElement(p)
 SAX.characters(
 , 2)
+SAX.endElement(p)
 SAX.startElement(ul)
 SAX.startElement(li)
 SAX.startElement(b)
@ -201,12 +220,16 @@ SAX.characters(DataManagement.FlatFile, 23)
 SAX.endElement(a)
 SAX.endElement(li)
 SAX.endElement(ul)
+SAX.startElement(p)
 SAX.characters(
 , 2)
+SAX.endElement(p)
 SAX.startElement(dir)
 SAX.endElement(dir)
+SAX.startElement(p)
 SAX.characters(
 , 2)
+SAX.endElement(p)
 SAX.startElement(ul)
 SAX.startElement(li)
 SAX.startElement(b)
@ -217,12 +240,16 @@ SAX.characters(DataManagement, 14)
 SAX.endElement(a)
 SAX.endElement(li)
 SAX.endElement(ul)
+SAX.startElement(p)
 SAX.characters(
 , 2)
+SAX.endElement(p)
 SAX.startElement(dir)
 SAX.endElement(dir)
+SAX.startElement(p)
 SAX.characters(
 , 2)
+SAX.endElement(p)
 SAX.endElement(body)
 SAX.endElement(html)
 SAX.ignorableWhitespace(
--- a/result/HTML/wired.html.sax
+++ b/result/HTML/wired.html.sax
@ -8,9 +8,11 @@ SAX.characters(Top Stories News from Wired Ne, 32)
 SAX.endElement(title)
 SAX.endElement(head)
 SAX.startElement(body, bgcolor='#FFFFFF', text='#000000', link='#333399', vlink='#660066', alink='#666699')
+SAX.startElement(p)
 SAX.characters(

 , 2)
+SAX.endElement(p)
 SAX.startElement(table, border='0', width='600', cellspacing='0', cellpadding='0')
 SAX.characters(
  , 3)
@ -315,6 +317,7 @@ SAX.endElement(tr)
 SAX.characters(
 , 1)
 SAX.endElement(table)
+SAX.startElement(p)
 SAX.characters(

 , 2)
@ -329,6 +332,7 @@ SAX.startElement(a, name='#')
 SAX.endElement(a)
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.startElement(table, border='0', width='600', cellspacing='0', cellpadding='0')
 SAX.characters(

@ -574,6 +578,7 @@ SAX.comment(
 SAX.characters(
 , 1)
 SAX.endElement(table)
+SAX.startElement(p)
 SAX.characters(
 , 1)
 SAX.comment( end WIRED NEWS header )
@ -584,6 +589,7 @@ SAX.comment( begin upper left side Navigation )
 SAX.characters(

 , 2)
+SAX.endElement(p)
 SAX.startElement(table, border='0', cellpadding='3', cellspacing='0', align='LEFT', bgcolor='#FFFFFF')
 SAX.characters(
  , 3)
@ -1465,6 +1471,7 @@ SAX.characters(

 , 2)
 SAX.endElement(table)
+SAX.startElement(p)
 SAX.characters(

 , 2)
@ -1475,6 +1482,7 @@ SAX.comment( CONTENT TABLE )
 SAX.characters(

 , 2)
+SAX.endElement(p)
 SAX.startElement(table, border='0', width='447', cellspacing='0', cellpadding='0', bordercolor='#66FF00')
 SAX.characters(
 , 2)
@ -2832,6 +2840,7 @@ SAX.endElement(tr)
 SAX.characters(
 , 1)
 SAX.endElement(table)
+SAX.startElement(p)
 SAX.characters(


@ -2840,9 +2849,14 @@ SAX.startElement(br)
 SAX.endElement(br)
 SAX.characters(
 , 1)
+SAX.endElement(p)
 SAX.endElement(body)
+SAX.startElement(body)
+SAX.startElement(p)
 SAX.characters(
 , 1)
+SAX.endElement(p)
+SAX.endElement(body)
 SAX.endElement(html)
 SAX.ignorableWhitespace(

--- a/tree.c
+++ b/tree.c
@ -426,10 +426,17 @@ xmlCreateIntSubset(xmlDocPtr doc, const xmlChar *name,
 	} else {
 	    xmlNodePtr prev;

-	    prev = doc->last;
-	    prev->next = (xmlNodePtr) cur;
-	    cur->prev = prev;
-	    doc->last = (xmlNodePtr) cur;
+	    if (doc->type == XML_HTML_DOCUMENT_NODE) {
+		prev = doc->children;
+		prev->prev = (xmlNodePtr) cur;
+		cur->next = prev;
+		doc->children = (xmlNodePtr) cur;
+	    } else {
+		prev = doc->last;
+		prev->next = (xmlNodePtr) cur;
+		cur->prev = prev;
+		doc->last = (xmlNodePtr) cur;
+	    }
 	}
    }
    return(cur);