filter html

gistfile1.js

filter = function(_html) {
        var qualifyURL = function(html) {
            html = html.replace(/href=\"([^\"]*)\"|href=\'([^\']*)\'/ig, function(url) {
                if(url == "" || url.search(/javascript:|vbscript:/) != -1){
                    return "";
                }
                try {
                    return 'href="' + qualify(url.split('href="')[1].split('"')[0]) + '"';
                } catch (e) {
                    return 'href="' + qualify(url.split("href='")[1].split("'")[0]) + '"';
                }
            });

            html = html.replace(/src=\"([^\"]*)\"|src=\'([^\']*)\'/ig, function(url) {
                try {
                    return 'src="' + qualify(url.split('src="')[1].split('"')[0]) + '"';
                } catch (e) {
                    return 'src="' + qualify(url.split("src='")[1].split("'")[0]) + '"';
                }
            });

            function qualify(url){
                var div = document.createElement('div');
                div.innerHTML = "<a></a>";
                div.firstChild.href = url; // Ensures that the href is properly escaped
                div.innerHTML = div.innerHTML; // Run the current innerHTML back through the parser
                return div.firstChild.href;
            }

            return html;
        },
        removeAttr = function(html) {
            var attr = ["mkmarkid", "style", "id", "class", "height", "width", "align", "\s{1,}on.[^\"|^\s|^<|^>]*?", "action", "background", "bgsound"],
                regex = "";
            for (var i = 0; i < attr.length; i++) {
                regex = new RegExp("(<\\w+[^>]+)(" + attr[i] + "=(('[^'>]+')|(\"[^\">]+\")|[^\\s>]+))", "ig");
                html = html.replace(regex, function($0,$1,$2) {
                    return $1;
                });
            }

            return html;
        },
        removeTag = function(html) {
            var tagRegex = /<(style|script|link|iframe|frame|frameset|noscript|head|html|applet|base|basefont|bgsound|blink|ilayer|layer|meta|object|embed)[^<>]*>[\s\S]*?.*?<\/\1>/igm;
            return html.replace(tagRegex, "");
        },
        removeWhite = function(html) {
            var whiteRegex = /\s{2,}|(\r|\n)+/g,//multi white or single newline
            whitegtRegex = /\s>/g; //" >"
            return html.replace(whiteRegex, " ...").replace(whitegtRegex, ">");
        };

        _html = removeTag(_html);
        _html = removeAttr(_html);
        //_html = removeWhite(_html); //this will remove the white space in code and pre
        _html = qualifyURL(_html);

        return _html;
    };