aggressive formatting for xml in php. Seems to solve most of the encoding and html entity problems. experimental

properXML

function properXML($str){
//return iconv("UTF-8", "ISO-8859-1//TRANSLIT", preg_replace("/&/", "and", html_entity_decode(strip_tags($str), ENT_COMPAT, 'UTF-8')));
return utf8_encode(preg_replace("/&/", "and", html_entity_decode(strip_tags($str), ENT_COMPAT, 'UTF-8')));
}