html = this.cleanHtml(html);
// fix up the special chars.. normaly like back quotes in word...
// however we do not want to do this with chinese..
- html = html.replace(/([\x80-\uffff])/g, function (a, b) {
- var cc = b.charCodeAt();
- if (
+ html = html.replace(/[\uD800-\uDBFF][\uDC00-\uDFFF]|[\u0080-\uFFFF]/g, function(match) {
+
+ var cc = match.charCodeAt();
+
+ // Get the character value, handling surrogate pairs
+ if (match.length == 2) {
+ // It's a surrogate pair, calculate the Unicode code point
+ var high = match.charCodeAt(0) - 0xD800;
+ var low = match.charCodeAt(1) - 0xDC00;
+ cc = (high * 0x400) + low + 0x10000;
+ } else if (
(cc >= 0x4E00 && cc < 0xA000 ) ||
(cc >= 0x3400 && cc < 0x4E00 ) ||
(cc >= 0xf900 && cc < 0xfb00 )
) {
- return b;
- }
- return "&#"+cc+";"
+ return match;
+ }
+
+ // No, use a numeric entity. Here we brazenly (and possibly mistakenly)
+ return "&#" + cc + ";";
+
+
});
+
+
+
if(this.owner.fireEvent('beforesync', this, html) !== false){
this.el.dom.value = html;
this.owner.fireEvent('sync', this, html);
insertTag : function(tg)
{
// could be a bit smarter... -> wrap the current selected tRoo..
- if (tg.toLowerCase() == 'span' || tg.toLowerCase() == 'code') {
+ if (tg.toLowerCase() == 'span' ||
+ tg.toLowerCase() == 'code' ||
+ tg.toLowerCase() == 'sup' ||
+ tg.toLowerCase() == 'sub'
+ ) {
range = this.createRange(this.getSelection());
var wrappingNode = this.doc.createElement(tg.toLowerCase());
var remove_keep_children= Roo.HtmlEditorCore.remove.indexOf(node.tagName.toLowerCase()) > -1;
+ // spans with no attributes - just remove them..
+ if ((!node.attributes || !node.attributes.length) && lcname == 'span') {
+ remove_keep_children = true;
+ }
+
// remove <a name=....> as rendering on yahoo mailer is borked with this.
// this will have to be flaged elsewhere - perhaps ablack=name... on the mailer..
}
if (!node.attributes || !node.attributes.length) {
+
+
+
+
this.cleanUpChildren(node);
return;
}
if (a.name == 'class') {
if (a.value.match(/^Mso/)) {
- node.className = '';
+ node.removeAttribute('class');
}
if (a.value.match(/^body$/)) {
- node.className = '';
+ node.removeAttribute('class');
}
continue;
}
node.nodeName == 'SPAN' &&
!node.hasAttributes() &&
node.childNodes.length == 1 &&
- node.firstChild.nodeName == "#text"
+ node.firstChild.nodeName == "#text"
) {
- var text = ' ' + node.innerHTML + ' ';
- var textNode = document.createTextNode(text);
+ var textNode = node.firstChild;
+ node.removeChild(textNode);
+ if (node.getAttribute('lang') != 'zh-CN') { // do not space pad on chinese characters..
+ node.parentNode.insertBefore(node.ownerDocument.createTextNode(" "), node);
+ }
node.parentNode.insertBefore(textNode, node);
+ if (node.getAttribute('lang') != 'zh-CN') { // do not space pad on chinese characters..
+ node.parentNode.insertBefore(node.ownerDocument.createTextNode(" ") , node);
+ }
node.parentNode.removeChild(node);
}
node.parentNode.removeChild(node);
return;
}
-
+ //Roo.log(node.tagName);
// remove - but keep children..
- if (node.tagName.toLowerCase().match(/^(meta|link|\\?xml:|st1:|o:|font)/)) {
+ if (node.tagName.toLowerCase().match(/^(meta|link|\\?xml:|st1:|o:|v:|font)/)) {
+ //Roo.log('-- removed');
while (node.childNodes.length) {
var cn = node.childNodes[0];
node.removeChild(cn);
node.parentNode.insertBefore(cn, node);
+ // move node to parent - and clean it..
+ this.cleanWord(cn);
}
node.parentNode.removeChild(node);
- this.iterateChildren(node, this.cleanWord);
+ /// no need to iterate chidlren = it's got none..
+ //this.iterateChildren(node, this.cleanWord);
return;
}
// clean styles