Index: openacs-4/packages/acs-templating/www/resources/xinha-nightly/plugins/GetHtml/get-html.js
===================================================================
RCS file: /usr/local/cvsroot/openacs-4/packages/acs-templating/www/resources/xinha-nightly/plugins/GetHtml/get-html.js,v
diff -u -r1.1 -r1.2
--- openacs-4/packages/acs-templating/www/resources/xinha-nightly/plugins/GetHtml/get-html.js	2 Nov 2005 08:56:01 -0000	1.1
+++ openacs-4/packages/acs-templating/www/resources/xinha-nightly/plugins/GetHtml/get-html.js	11 Nov 2005 20:32:43 -0000	1.2
@@ -12,14 +12,15 @@
   *   - Returns correct code for Flash objects and scripts
   *   - Formats html in an indented, readable format in html mode
   *   - Preserves script and pre formatting
+  *   - Preserves formatting in comments
   *   - Removes contenteditable from body tag in full-page mode
   *   - Supports only7BitPrintablesInURLs config option
   *   - Supports htmlRemoveTags config option
   */
   
 function GetHtml(editor) {
     this.editor = editor;
-};
+}
 
 GetHtml._pluginInfo = {
 	name          : "GetHtml",
@@ -44,12 +45,13 @@
 /*11*/  new RegExp().compile(/\s+([^=\s]+)(="[^"]+")/g),// lowercase attribute names
 /*12*/  new RegExp().compile(/(\S*\s*=\s*)?contenteditable[^=>]*(=\s*[^>\s\/]*)?/gi),//strip contenteditable
 /*13*/  new RegExp().compile(/((href|src)=")([^\s]*)"/g), //find href and src for stripBaseHref()
-/*14*/  new RegExp().compile(/<\/?(div|p|h[1-6]|table|tr|td|th|ul|ol|li|blockquote|object|br|hr|img|embed|param|pre|script|html|head|body)[^>]*>/g),
-/*15*/  new RegExp().compile(/<\/(div|p|h[1-6]|table|tr|td|th|ul|ol|li|blockquote|object|html|head|body)( [^>]*)?>/g),//blocklevel closing tag
-/*16*/  new RegExp().compile(/<(div|p|h[1-6]|table|tr|td|th|ul|ol|li|blockquote|object|html|head|body)( [^>]*)?>/g),//blocklevel opening tag
-/*17*/  new RegExp().compile(/<(br|hr|img|embed|param|pre|script)[^>]*>/g),//singlet tag
+/*14*/  new RegExp().compile(/<\/?(div|p|h[1-6]|table|tr|td|th|ul|ol|li|blockquote|object|br|hr|img|embed|param|pre|script|html|head|body|meta|link|title)[^>]*>/g),
+/*15*/  new RegExp().compile(/<\/(div|p|h[1-6]|table|tr|td|th|ul|ol|li|blockquote|object|html|head|body|script)( [^>]*)?>/g),//blocklevel closing tag
+/*16*/  new RegExp().compile(/<(div|p|h[1-6]|table|tr|td|th|ul|ol|li|blockquote|object|html|head|body|script)( [^>]*)?>/g),//blocklevel opening tag
+/*17*/  new RegExp().compile(/<(br|hr|img|embed|param|pre|meta|link|title)[^>]*>/g),//singlet tag
 /*18*/  new RegExp().compile(/(^|<\/(pre|script)>)(\s|[^\s])*?(<(pre|script)[^>]*>|$)/g),//exclude content between pre and script tags
-/*19*/  new RegExp().compile(/(<pre[^>]*>)(\s|[^\s])*?(<\/pre>)/g)//find content inside pre tags
+/*19*/  new RegExp().compile(/(<pre[^>]*>)(\s|[^\s])*?(<\/pre>)/g),//find content inside pre tags
+/*20*/  new RegExp().compile(/(^|<!--(\s|\S)*?-->)((\s|\S)*?)(?=<!--(\s|\S)*?-->|$)/g)//exclude comments
 ];
 
 /** 
@@ -73,7 +75,7 @@
 		replace(c[10], ' ');//trim extra whitespace
 	if(HTMLArea.is_ie && c[13].test(sHtml)) {//
 		sHtml = sHtml.replace(c[13],'$1'+this.stripBaseURL(RegExp.$3)+'"');
-	};
+	}
 	if(this.config.only7BitPrintablesInURLs && c[13].test(sHtml)) {
 	  sHtml = sHtml.replace(c[13], '$1'+RegExp.$3.replace(/([^!-~]+)/g,function(chr){return escape(chr);})+'"');
 	}
@@ -91,7 +93,8 @@
 	if(HTMLArea.is_gecko) { //moz changes returns into <br> inside <pre> tags
 		s = s.replace(c[19], function(str){return str.replace(/<br \/>/g,"\n")});
 	}
-	s = s.replace(c[18], function(string) {
+	s = s.replace(c[20], function(st,$1,$2,$3) { //exclude comments
+	  strn = $3.replace(c[18], function(string) { //skip pre and script tags
 		string = string.replace(/[\n\r]/gi, " ").replace(/\s+/gi," ").replace(c[14], function(str) {
 			if (str.match(c[16])) {
 				var s = "\n" + HTMLArea.__sindent + str;
@@ -114,6 +117,7 @@
 			return str; // this won't actually happen
 		});
 		return string;
+      });return $1 + strn;
     });
     if (s.charAt(0) == "\n") {
         return s.substring(1, s.length);
@@ -133,32 +137,49 @@
 	    for (j = temp.nextSibling; j; j = j.nextSibling) { 
 	    		temp.appendChild(j.cloneNode(true));
 	    }
-	    html += temp.innerHTML.replace(/<[^\?][^>]*>/gi, function(tag){return editor.cleanHTML(tag)});
+	    html += temp.innerHTML.replace(/<[^\?!][^>]*>/gi, function(tag){return editor.cleanHTML(tag)});
 
 	} else {
 
-	var root_tag = (root.nodeType == 1) ? root.tagName.toLowerCase() : ''; 
-	if (outputRoot) { //only happens with <html> tag in fullpage mode
-		html += "<" + root_tag + ">";
-	}
-
-	//pass tags to cleanHTML() one at a time; ignore php tags
-	//includes support for htmlRemoveTags config option
-	html += editor.getInnerHTML().replace(/<[^\?][^>]*>/gi, function(tag){
-		if(!(editor.config.htmlRemoveTags && editor.config.htmlRemoveTags.test(tag.replace(/<([^\s>\/]+)/,'$1'))))
-		return editor.cleanHTML(tag);
-		else return ''});
-	//IE drops  all </li> tags in a list except the last one
-	if(HTMLArea.is_ie) {
-		html = html.replace(/<li( [^>]*)?>/g,'</li><li$1>').
-			replace(/(<(ul|ol)[^>]*>)[\s\n]*<\/li>/g, '$1').
-			replace(/<\/li>([\s\n]*<\/li>)+/g, '<\/li>');
-	}
-	if (outputRoot) {
-		html += "</" + root_tag + ">";
-	}
-	html = HTMLArea.indent(html);
-}
+		var root_tag = (root.nodeType == 1) ? root.tagName.toLowerCase() : ''; 
+		if (outputRoot) { //only happens with <html> tag in fullpage mode
+			html += "<" + root_tag;
+			var attrs = root.attributes; // strangely, this doesn't work in moz
+			for (i = 0; i < attrs.length; ++i) {
+				var a = attrs.item(i);
+				if (!a.specified) {
+				  continue;
+				}
+				var name = a.nodeName.toLowerCase();
+				var value = a.nodeValue;
+				html += " " + name + '="' + value + '"';
+			}
+			html += ">";
+		}
+		if(root_tag == "html") {
+			innerhtml = editor._doc.documentElement.innerHTML;
+		} else {
+			innerhtml = editor._doc.body.innerHTML;
+		}
+		//pass tags to cleanHTML() one at a time; ignore comments and php tags
+		//includes support for htmlRemoveTags config option
+		html += innerhtml.replace(/<[^\?!][^>]*>/gi, function(tag){
+			if(!(editor.config.htmlRemoveTags && editor.config.htmlRemoveTags.test(tag.replace(/<([^\s>\/]+)/,'$1'))))
+			return editor.cleanHTML(tag);
+			else return ''});
+		//IE drops  all </li> tags in a list except the last one
+		if(HTMLArea.is_ie) {
+			html = html.replace(/<li( [^>]*)?>/g,'</li><li$1>').
+				replace(/(<(ul|ol)[^>]*>)[\s\n]*<\/li>/g, '$1').
+				replace(/<\/li>([\s\n]*<\/li>)+/g, '<\/li>');
+		}
+		if(HTMLArea.is_gecko)
+			html = html.replace(/(.*)<br \/>\n$/, '$1'); //strip trailing <br> added by moz
+		if (outputRoot) {
+			html += "</" + root_tag + ">";
+		}
+		html = HTMLArea.indent(html);
+	};
 //	html = HTMLArea.htmlEncode(html);
 
 	return html;