// ==UserScript==
// @run-at document-start
// @name           General URL Cleaner
// @namespace      
// @description    Cleans URL's from various popular sites. Also, makes sure the sites are using HTTPS.
// @include        /^https?://[a-z]*.google.(com|ca|ac|ad|ae|al|am|us|as|at|az|ba|be|bf|bg|bi|bj|bs|btby|cc|cd|cf|cat|cg)\/.*$/
// @include        /^https?://[a-z]*.google.(ch|ci|cl|cm|cn|cv|cz|de|dj|dk|dm|dz|ee|es|fi|fm|frga|ge|gg|gl|gm|gp|gr|gy|hn)\/.*$/
// @include        /^https?://[a-z]*.google.(iq|ie|im|io|is|it|je|jo|ki|kg|kz|hr|ht|hu|ir|la|li|lk|lt|lu|lv|md|me|mg|mk|ml)\/.*$/
// @include        /^https?://[a-z]*.google.(mn|ms|mu|mv|mw|ne|nl|nonr|nu|pl|pn|ps|vg|pt|ro|rs|ru|rw|sc|se|sh|si|sk|sn|sm)\/.*$/
// @include        /^https?://[a-z]*.google.(so|st|td|tg|tk|tl|tm|to|tn|tt|vu|ws|com.(sg|sl|sv|tj|tn|tr|tw|ua|uy|vc|vn))\/.*$/
// @include        /^https?://[a-z]*.google.com.(au|af|ag|ai|ar|bh|bn|bo|br|bz|kh|co|cu|cy|do|ec|eg|et|fj|gh|gi|gt|hk|jm)\/.*$/
// @include        /^https?://[a-z]*.google.com.(kw|lb|lc|ly|mm|mt|mx|my|na|nf|ng|ni|np|om|pa|pe|ph|pk|pg|pr|py|qa|sa|sb)\/.*$/
// @include        /^https?://[a-z]*.google.co.(uk|nz|ao|ar|bw|ck|cr|id|il|in|jp|ke|kr|ls|ma|mz|th|tz|ug|uz|ve|vi|za|zm|zw)\/.*$/
// @include        /^https?://[a-z]*.amazon.(cn|in|co.jp|fr|de|it|nl|es|co.uk|ca|com.mx|com|com.au|com.br)\/.*$/
// @include        /^https?://[a-z]*.newegg.(com|ca|cn)\/.*$/
// @include        /^https?://[a-z]*.ebay.(com.au|at|be|ca|fr|de|com.hk|in|ie|co.il|it|com.my|nl|co.za|ph|pl|com.sg|co.za|es|ch|co.th|co.uk|com|vn)\/.*$/
// @include        /^https?://[a-z]*.bing.com\/.*$/
// @include        /^https?://[a-z]*.youtube.com\/.*$/
// @include        /^https?://[a-z]*.dealtime.com\/.*$/
// @exclude        https://apis.google.com/*
// @exclude        https://www.google.com/recaptcha/api2/*
// @version        1.9.2.2
// @license        GPL version 3 or any later version; http://www.gnu.org/copyleft/gpl.html
// @downloadURL none
// ==/UserScript==

// compile these regexes beforehand to improve efficiency
// These regex's are for checking which site the URL belongs to
var bing = new RegExp(/^https?:\/\/www\.bing\.(.+?)\/search\?/);
var google = new RegExp(/^https?:\/\/[a-z]*\.google\.(.+?)\/[a-z]*\?/);
var googleImageRedirect = new RegExp(/^https?:\/\/www\.google\.(.+?)\/url\?.*\&url\=/);
var youtube = new RegExp(/^https?:\/\/www\.youtube\.com\/watch/);
var youtubeRedirectLink = new RegExp(/^https?:\/\/www.youtube.com\/redirect\?q\=/);
var ebay = new RegExp(/^https?:\/\/www\.ebay\.(.+?)\/itm/);
var ebaySearch = new RegExp(/^https?:\/\/www\.ebay\.(.+?)\/sch\//);
var amazondp = new RegExp(/^https?:\/\/www\.amazon\..*\/dp\//);
var amazongp = new RegExp(/^https?:\/\/www\.amazon\..*\/gp\/product\//);
var newegg = new RegExp(/^http:\/\/www\.newegg\.(com|ca)\/Product\/Product\.aspx/);
var dealtime = new RegExp(/http:\/\/stat\.dealtime\.com\/DealFrame\/DealFrame\.cmp\?/);
//  These regex's are for replacing parts of the URL
var utmParameters = new RegExp(/((\?|\&|)utm_(source|medium|campaign)\=[^&]*|\&amp\;)/g);
var googleSearchParameters = new RegExp(
    '\&(aqs|es_sm|channel|tab|num|hl|safe|tbo|sclient|sourceid|spell|site|sa|ei|client|complete|as_qdr|um|sa|tab|' +
    'authuser|rlz|cad|rct|ved|usg|source|oe|oq|ie|dpr|gs_l|ved|tbas|sei|biw|bih|gpsrc|gfe_rd|gws_rd)\=[^&]*', 'gi');

// Clean the current page URL
var newPageUrl = cleanUrl(document.URL);
if (newPageUrl != document.URL) location.replace(newPageUrl);

// Cleans links on the page
var links = document.links;
// don't do anything with links that are blank, javascript, email addresses, data
var excludeLinks = new RegExp(/(^$|^javascript\:|^mailto\:|^data\:)/);

if (google.test(newPageUrl)) {
    document.addEventListener("DOMContentLoaded", cleanGooglePageLinks, false);
    window.onhashchange = googleInstant;
}
else {
    document.addEventListener("DOMContentLoaded", cleanPageLinks, false);
}

// Standard link cleaning function
function cleanPageLinks() {
    for (var i = links.length; i--;) {
        if (excludeLinks.test(links[i].href)) continue; // Links to skip
        links[i].href = cleanUrl(links[i].href);        // Standard link cleaning
    }
    // We don't need to keep the event listener running
    this.removeEventListener('DOMContentLoaded', cleanPageLinks, false);
}

// Google search results link cleaning function
function cleanGooglePageLinks() {
    for (var i = links.length; i--;) {
        if (excludeLinks.test(links[i].href)) continue; // Links to skip
        links[i].removeAttribute('onmousedown');        // Remove search results redirection
        links[i].href = cleanUrl(links[i].href);        // Standard link cleaning
    }
    // We don't need to keep event listener running
    this.removeEventListener('DOMContentLoaded', cleanGooglePageLinks, false);
}

// Google Instant document URL cleaning - if the search terms change, remove the extra stuff.
function googleInstant() {
    // Don't rewrite anything if an image is clicked in image searches
    if (!document.URL.includes('#imgrc=')) {
        // The string after the hash, containing the new search terms
        var newSearchString = String(document.URL.match(/\#.*/)).replace(/^\#/,'');
        
        // Remake the full URL with only the new search terms
        var newSearchUrl = String(document.URL.replace(/search\?.*/, 'search?' + newSearchString));
        location.replace(newSearchUrl);
    }
}

// Main function for cleaning the url's
function cleanUrl(oldurl) {
    var newurl = oldurl;
    switch(true) {
        case googleImageRedirect.test(oldurl):
            newurl = decodeURIComponent(oldurl.replace(googleImageRedirect,'').replace(/\&(psig|ei|bvm)\=.*$/g,''));
            break;
        case google.test(oldurl):
                     // temporarily put an "&" after the "?" so that the search parameters regex will always work
            newurl = oldurl.replace('?','?&')
                           .replace(googleSearchParameters,'')
                           .replace('?&','?')
                           .replace(/^http\:/,'https:'); // always use https
            break;
        case bing.test(oldurl):
            newurl = oldurl.replace('?','?&')
                           .replace(/\&(go|qs|form|FORM|filt|pq|sc|sp|sk|qpvt)\=[^&]*/g,'')
                           .replace('?&','?')
                           .replace(/^http\:/,'https:');
            break;
        case youtube.test(oldurl):
            newurl = 'https://www.youtube.com/watch?' + oldurl.match(/v\=[^&]*/);
            break;
        case youtubeRedirectLink.test(oldurl):
            newurl = decodeURIComponent(oldurl.replace(youtubeRedirectLink,'').replace(/\&redir_token\=.*/,''));
            break;
        case ebay.test(oldurl):
            // the split gets the domain name. Should be more efficient than a regex.
            newurl = 'http://' + oldurl.split('/')[2] + '/itm' + oldurl.match(/\/[0-9]{11,13}[^#?&\/]/);
            break;
        case ebaySearch.test(oldurl):
            newurl = oldurl.replace('?','?&')
                           .replace(/\&(\_osacat|\_odkw|\_from|rt|\_trksid|\_sacat)\=[^&]*/g,'')
                           .replace('?&','?');
            break;
        case amazondp.test(oldurl):
            newurl = 'http://' + oldurl.split('/')[2] + oldurl.match(/\/dp\/[A-Z0-9]{10}\/?/);
            break;
        case amazongp.test(oldurl):
            newurl = 'http://' + oldurl.split('/')[2] + oldurl.match(/\/gp\/product\/[A-Z0-9]{10}\/?/);
            break;
        case newegg.test(oldurl):
            newurl = 'http://' + oldurl.split('/')[2] + oldurl.match(/\/Product\/Product\.aspx\?Item\=[^&]*/);
            break;
        case dealtime.test(oldurl):
            newurl = decodeURIComponent(oldurl.replace(/.*\&url\=/,'').replace(/(\%26|)\&linkin_id\=.*$/,'')).replace(/\&(url|partner)\=[^&]*/g,'');
            break;
        default:
            break;
    }
    newurl = newurl.replace(utmParameters,'');
    return newurl;
}