X-Git-Url: https://git.ao2.it/GM_direct_download_links.git/blobdiff_plain/4912dcb94c0ab805a92c012c8da9f86246c7eeba..ca6cfbc09afa37b5f618e481aadf9ed2aafc2fd4:/direct_download_links.user.js diff --git a/direct_download_links.user.js b/direct_download_links.user.js index 0a9aac5..e1f17f3 100644 --- a/direct_download_links.user.js +++ b/direct_download_links.user.js @@ -1,6 +1,6 @@ -// direct_download_links - Add direct download links for (video) files -// version 0.2 -// 2011-11-14 +// direct_download_links - Add direct download links +// version 0.3 +// 2011-12-23 // Copyright (C) 2011 Antonio Ospite // Released under the GPL license // http://www.gnu.org/copyleft/gpl.html @@ -22,46 +22,103 @@ // ==UserScript== // @name Direct Download Links // @namespace http://git.ao2.it/GM_direct_download_links.git -// @description Add direct download links for (video) files +// @description Add direct download links // @include http://video.repubblica.it/* // @include http://tv.repubblica.it/* // @include http://trovacinema.repubblica.it/* // @include http://www.kataweb.it/tvzap/* +// @include http://www.rai.tv/* +// @include http://soundcloud.com/* // ==/UserScript== +// /* * TODO: * - find a way to use the same string as in the @include lines to match the - * current window.location + * current window.location. Look for something like GM_testUrl() which builds + * the regexp starting from a glob line. * - use xpath instead of regexp like in http://a32.me/2009/11/greasemonkey/ * - use jquery, like shown in http://a32.me/2009/11/greasemonkey/ + * - Support the "download" attribute for anchors: + * http://www.whatwg.org/specs/web-apps/current-work/multipage/links.html#downloading-resources */ +/* Fields supported by the "site" object. + * + * Manadatory fields: + * pageURL: the URL of the page we are modifying + * urlContainer: the element containing the URL to link + * urlRegexp: the regular expression for finding the URL, the first + * sub-pattern is taken as the URL + * linkDest: the element where to place the Direct Download link + * + * + * Optional fields: + * + * initCommand: a function called before the regExp is matched, this can + * be useful in cases when some action needs to be done in + * order to make the element containing the regExp be actually + * rendered. It must accept a 'site' parameter. + * + * onEvent: used to delay the urlRegexp matching to a certain event like + * 'DOMNodeInserted' useful when the URL is added by some javascript + * library. It has two fields: + * + * evt: the event we want to wait for (e.g. 'DOMNodeInserted') + * + * targetElement: the element in the event handler we want the + * urlRegexp is performed on. + * + * processURL: a function to process the URL before adding the Direct + * Downdload Link to the page, it must accept a 'site' and a + * 'URL' parameters and dispatch the UrlFetched to pass the + * modified URL to _add_link(). + * + */ var supported_sites = [ { locationRegexp: /^http:\/\/video\.repubblica\.it\/.*$/, - fileElem: 'contA', - fileRegexp: /'pcUrl', '((http|mms):\/\/[^']*)'/, + urlContainer: 'contA', + urlRegexp: /'pcUrl', '((http|mms):\/\/[^']*)'/, linkDest: 'contA', }, { locationRegexp: /^http:\/\/tv\.repubblica\.it\/.*$/, - fileElem: 'boxPlayer', - fileRegexp: /'pcUrl', '((http|mms):\/\/[^']*)'/, + urlContainer: 'boxPlayer', + urlRegexp: /'pcUrl', '((http|mms):\/\/[^']*)'/, linkDest: 'box_embed', }, { locationRegexp: /^http:\/\/trovacinema\.repubblica\.it\/.*$/, - fileElem: 'col-center', - fileRegexp: /'flvUrl', '((http|mms):\/\/[^']*)'/, + urlContainer: 'col-center', + urlRegexp: /'flvUrl', '((http|mms):\/\/[^']*)'/, linkDest: 'col-center', }, { locationRegexp: /^http:\/\/www\.kataweb\.it\/tvzap\/.*$/, - fileElem: 'tvzap_video', - fileRegexp: /'pcUrl', '((http|mms):\/\/[^']*)'/, + urlContainer: 'tvzap_video', + urlRegexp: /'pcUrl', '((http|mms):\/\/[^']*)'/, linkDest: 'playerCont', }, + { + locationRegexp: /^http:\/\/www\.rai\.tv\/.*$/, + initCommand: function(site) { + unsafeWindow.Silverlight.isInstalled = function(version) { + return true; + }; + }, + urlContainer: 'Player', + urlRegexp: /mediaUri=(http:\/\/[^,]*)/, + onEvent: { evt: 'DOMNodeInserted', targetElement: 'object' }, + processURL: _rai_get_actual_url, + linkDest: 'Player', + }, + { + locationRegexp: /^http:\/\/soundcloud.com\/.*$/, + urlContainer: 'main-content-inner', + urlRegexp: /"streamUrl":"([^"]*)"/, + linkDest: 'main-content-inner', + }, ]; /* Apply different rules to different sites */ @@ -70,58 +127,153 @@ for (i = 0; i < supported_sites.length; i++) { var result = window.location.href.match(site.locationRegexp); if (result) { - var ret = direct_download_link_add(window.location.href, site.fileElem, site.fileRegexp, site.linkDest); - if (!ret) { - alert('Cannot add the link'); + if (site.initCommand) { + site.initCommand(site); } + direct_download_link_add(window.location.href, site); } } /** - * Add a Direct Download link on the page for the specified file + * Add a Direct Download link on the page for the specified URL * - * @param pageURL: the URL of the page we are modifying - * @param fileElem: the element containing the file URL - * @param fileRegexp: the regular expression for finding the file, the first - * sub-pattern is taken as the file URL - * @param linkDest: the element where to place the Direct Download link + * @param: a 'site' object described above. * * @return: null on error, true on success */ -function direct_download_link_add(pageURL, fileElem, fileRegexp, linkDest) { - var element = document.getElementById(fileElem); +function direct_download_link_add(pageURL, site) { + var element = document.getElementById(site.urlContainer); if (!element) { - alert('DirectDL (' + pageURL + '): Cannot find the element containing the file URL.'); + DDL_log('DirectDL (' + site.pageURL + '): Cannot find the element ' + site.urlContainer + ' containing the URL.'); return null; } - var content = element.textContent; + document.addEventListener('UrlFetched', _add_link, true); + + // This is used for sites adding the URL to the DOM after DOMContentLoaded, + // for example by some javascript library (like Silverlight.js on rai.tv). + if (site.onEvent) { + element.addEventListener(site.onEvent.evt, function(e) { + if (site.onEvent.targetElement && + e.target.tagName.toLowerCase() != site.onEvent.targetElement) { + DDL_log('DirectDL (' + site.pageURL + '): skipping element ' + e.target.tagName); + return; + } + _get_URL(site, element); + }, false); + return; + } + + _get_URL(site, element); +} + +function _get_URL(site, element) { + var content = element.innerHTML; if (!content) { - alert('DirectDL (' + pageURL + '): textContent is null, cannot fild file URL.'); + DDL_log('DirectDL (' + site.pageURL + '): content is null, cannot find URL.'); + return; } - var fileURL = content.match(fileRegexp)[1]; - if (!fileURL) { - alert('DirectDL (' + pageURL + '): file URL not found, check the fileRegexp'); - return null; + var matches = content.match(site.urlRegexp); + if (!matches || matches.length < 2 || !matches[1]) { + DDL_log('DirectDL (' + site.pageURL + '): URL not found, check the urlRegexp'); + return; + } + var URL = matches[1]; + if (!URL) { + DDL_log('DirectDL (' + site.pageURL + '): cannot get the URL.'); + return; } - var links = document.getElementById(linkDest); - if (!links) { - alert('DirectDl (' + pageURL + '): Cannot add the direct download link.'); - return null; + if (site.processURL) { + site.processURL(site, URL); + return; + } + + var evt = document.createEvent('Event'); + evt.initEvent('UrlFetched', true, true); + evt.site = site; + evt.URL = URL; + document.dispatchEvent(evt); +} + +function _add_link(e) { + var site = e.site; + var URL = e.URL;; + + var destination = document.getElementById(site.linkDest); + if (!destination) { + DDL_log('DirectDl (' + site.pageURL + '): Cannot add the direct download link.'); + return; + } + + // Check if we added the link already, if so just update the href attribute. + // This is useful when _get_URL() is called on async events. + var download_link = document.getElementById('GM_direct_downaload_link'); + if (download_link) { + download_link.setAttribute('href', URL); + } else { + download_link = document.createElement('a'); + download_link.textContent = 'Direct Link'; + download_link.setAttribute('id', 'GM_direct_downaload_link'); + download_link.setAttribute('href', URL); + var style = 'background-color: white; color: blue;'; + style += ' border: 2px solid red;' + style += ' float: right; font-size: large;'; + style += ' padding: .5em; margin: 1em;' + style += ' position: relative; z-index: 1000;' + download_link.setAttribute('style', style); + + destination.insertBefore(download_link, destination.firstChild); + } +} + +function DDL_log(message) { + var debug = false; + if (debug) { + alert(message) + } else { + GM_log(message); + } +} + +function _rai_get_actual_url(site, URL) { + + // SmoothStreaming manifest files get added without processing, for now: + if (URL.match(/.*\.csm$/)) { + var evt = document.createEvent('Event'); + evt.initEvent('UrlFetched', true, true); + evt.site = site; + evt.URL = URL; + document.dispatchEvent(evt); + return; } - var download_link = document.createElement('a'); - download_link.textContent = 'Direct Link'; - download_link.setAttribute('href', fileURL); - var style = 'background-color: white; color: blue;'; - style += ' border: 2px solid red;' - style += ' float: right; font-size: large;'; - style += ' padding: .5em; margin: 1em;' - download_link.setAttribute('style', style); + // http://www.neaveru.com/wordpress/index.php/2008/05/09/greasemonkey-bug-domnodeinserted-event-doesnt-allow-gm_xmlhttprequest/ + setTimeout( function() { + GM_xmlhttpRequest({ + method: "GET", + // XXX A custom header. This is the "clever" trick Rai uses to ensure + // the content is accessed by www.rai.tv only... + headers: {'viaurl': 'www.rai.tv'}, + url: URL, + onload: function(response) { + text = response.responseText; + text = text.replace(/&/g, '&') + parser = new DOMParser(); + xmlDoc = parser.parseFromString(text, "text/xml"); - links.insertBefore(download_link, links.firstChild); + ref = xmlDoc.getElementsByTagName('REF'); + if (ref.length > 0) { + href = ref[0].getAttribute('HREF');; - return true; + var evt = document.createEvent('Event'); + evt.initEvent('UrlFetched', true, true); + evt.site = site; + evt.URL = href; + document.dispatchEvent(evt); + } + } + }); + }, 0); }