sync/lib/get-info.js

859 lines
28 KiB
JavaScript
Raw Normal View History

2013-03-24 02:28:20 +00:00
/*
The MIT License (MIT)
Copyright (c) 2013 Calvin Montgomery
2013-05-02 15:28:29 +00:00
2013-03-24 02:28:20 +00:00
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
2013-05-02 15:28:29 +00:00
2013-03-24 02:28:20 +00:00
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
2013-05-02 15:28:29 +00:00
2013-03-24 02:28:20 +00:00
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
*/
2013-02-16 17:19:59 +00:00
var http = require("http");
var https = require("https");
var domain = require("domain");
2013-03-27 19:28:51 +00:00
var Logger = require("./logger.js");
2013-04-03 03:56:44 +00:00
var Media = require("./media.js").Media;
var CustomEmbedFilter = require("./customembed").filter;
2013-10-11 21:31:40 +00:00
var Server = require("./server");
2014-01-22 23:11:26 +00:00
var Config = require("./config");
2013-10-11 21:31:40 +00:00
var urlRetrieve = function (transport, options, callback) {
// Catch any errors that crop up along the way of the request
// in order to prevent them from reaching the global handler.
// This should cut down on needing to restart the server
var d = domain.create();
d.on("error", function (err) {
2014-03-03 23:55:55 +00:00
Logger.errlog.log(err.stack);
2013-10-11 21:31:40 +00:00
Logger.errlog.log("urlRetrieve failed: " + err);
Logger.errlog.log("Request was: " + options.host + options.path);
callback(503, err);
});
d.run(function () {
var req = transport.request(options, function (res) {
var buffer = "";
res.setEncoding("utf-8");
res.on("data", function (chunk) {
buffer += chunk;
});
res.on("end", function () {
callback(res.statusCode, buffer);
});
});
2013-08-03 14:52:24 +00:00
2013-10-11 21:31:40 +00:00
req.end();
});
};
var Getters = {
/* youtube.com */
yt: function (id, callback) {
var sv = Server.getServer();
2014-01-22 23:11:26 +00:00
/*
2013-10-11 21:31:40 +00:00
if (sv.cfg["enable-ytv3"] && sv.cfg["ytv3apikey"]) {
Getters["ytv3"](id, callback);
return;
}
2014-01-22 23:11:26 +00:00
*/
2013-08-01 03:26:11 +00:00
2013-10-11 21:31:40 +00:00
var m = id.match(/([\w-]+)/);
if (m) {
id = m[1];
} else {
callback("Invalid ID", null);
return;
}
2013-10-11 21:31:40 +00:00
var options = {
host: "gdata.youtube.com",
port: 443,
path: "/feeds/api/videos/" + id + "?v=2&alt=json",
method: "GET",
dataType: "jsonp",
timeout: 1000
};
2014-01-22 23:11:26 +00:00
if(Config.get("youtube-v2-key")) {
2013-10-11 21:31:40 +00:00
options.headers = {
2014-01-22 23:11:26 +00:00
"X-Gdata-Key": "key=" + Config.get("youtube-v2-key")
};
2013-10-11 21:31:40 +00:00
}
2013-10-11 21:31:40 +00:00
urlRetrieve(https, options, function (status, data) {
if(status === 404) {
callback("Video not found", null);
return;
} else if(status === 403) {
callback("Private video", null);
return;
} else if(status === 400) {
callback("Invalid video", null);
return;
} else if(status === 503) {
callback("API failure", null);
return;
} else if(status !== 200) {
callback("HTTP " + status, null);
return;
}
2013-10-11 21:31:40 +00:00
var buffer = data;
try {
data = JSON.parse(data);
if (data.entry.yt$accessControl) {
var ac = data.entry.yt$accessControl;
for (var i = 0; i < ac.length; i++) {
if (ac[i].action === "embed") {
if (ac[i].permission === "denied") {
callback("Embedding disabled", null);
return;
}
2013-10-11 21:31:40 +00:00
break;
}
}
2013-10-11 21:31:40 +00:00
}
2013-11-11 04:26:30 +00:00
2013-10-11 21:31:40 +00:00
var seconds = data.entry.media$group.yt$duration.seconds;
var title = data.entry.title.$t;
var media = new Media(id, title, seconds, "yt");
2013-11-11 04:26:30 +00:00
if (data.entry.media$group.media$restriction) {
var rest = data.entry.media$group.media$restriction;
if (rest.length > 0) {
if (rest[0].relationship === "deny") {
media.restricted = rest[0].$t;
}
}
}
2013-10-11 21:31:40 +00:00
callback(false, media);
} catch(e) {
// Gdata version 2 has the rather silly habit of
// returning error codes in XML when I explicitly asked
// for JSON
var m = buffer.match(/<internalReason>([^<]+)<\/internalReason>/);
if(m === null)
m = buffer.match(/<code>([^<]+)<\/code>/);
var err = e;
if(m) {
if(m[1] === "too_many_recent_calls") {
err = "YouTube is throttling the server right "+
"now for making too many requests. "+
"Please try again in a moment.";
} else {
err = m[1];
}
}
2013-10-11 21:31:40 +00:00
callback(err, null);
}
});
},
/* youtube.com API v3 (requires API key) */
2014-01-22 23:11:26 +00:00
// DEPRECATED
2013-10-11 21:31:40 +00:00
ytv3: function (id, callback) {
var sv = Server.getServer();
var m = id.match(/([\w-]+)/);
if (m) {
id = m[1];
} else {
callback("Invalid ID", null);
return;
}
var params = [
"part=" + encodeURIComponent("id,snippet,contentDetails"),
"id=" + id,
"key=" + sv.cfg["ytapikey"]
].join("&");
var options = {
host: "www.googleapis.com",
port: 443,
path: "/youtube/v3/videos?" + params,
method: "GET",
dataType: "jsonp",
timeout: 1000
};
urlRetrieve(https, options, function (status, data) {
if(status !== 200) {
callback("YTv3: HTTP " + status, null);
return;
}
2013-08-03 14:52:24 +00:00
2013-10-11 21:31:40 +00:00
try {
data = JSON.parse(data);
// I am a bit disappointed that the API v3 just doesn't
// return anything in any error case
if(data.pageInfo.totalResults !== 1) {
callback("Video not found", null);
2013-08-03 14:52:24 +00:00
return;
}
2013-10-11 21:31:40 +00:00
var vid = data.items[0];
var title = vid.snippet.title;
// No, it's not possible to get a number representing
// the video length. Instead, I get a time of the format
// PT#M#S which represents
// "Period of Time" # Minutes, # Seconds
var m = vid.contentDetails.duration.match(/PT(\d+)M(\d+)S/);
var seconds = parseInt(m[1]) * 60 + parseInt(m[2]);
var media = new Media(id, title, seconds, "yt");
callback(false, media);
} catch(e) {
callback(e, null);
}
2013-10-11 21:31:40 +00:00
});
},
/* youtube.com playlists */
yp: function (id, callback, url) {
var sv = Server.getServer();
var m = id.match(/([\w-]+)/);
if (m) {
id = m[1];
} else {
callback("Invalid ID", null);
return;
}
var path = "/feeds/api/playlists/" + id + "?v=2&alt=json";
// YouTube only returns 25 at a time, so I have to keep asking
// for more with the URL they give me
if(url !== undefined) {
path = "/" + url.split("gdata.youtube.com")[1];
}
var options = {
host: "gdata.youtube.com",
port: 443,
path: path,
method: "GET",
dataType: "jsonp",
timeout: 1000
};
2014-01-22 23:11:26 +00:00
if(Config.get("youtube-v2-key")) {
2013-10-11 21:31:40 +00:00
options.headers = {
2014-01-22 23:11:26 +00:00
"X-Gdata-Key": "key=" + Config.get("youtube-v2-key")
};
2013-10-11 21:31:40 +00:00
}
2013-10-11 21:31:40 +00:00
urlRetrieve(https, options, function (status, data) {
if(status === 404) {
callback("Playlist not found", null);
return;
} else if(status === 403) {
callback("Playlist is private", null);
return;
} else if(status === 503) {
callback("API failure", null);
return;
} else if(status !== 200) {
callback("YTPlaylist HTTP " + status, null);
}
2013-10-11 21:31:40 +00:00
try {
data = JSON.parse(data);
var vids = [];
for(var i in data.feed.entry) {
try {
var item = data.feed.entry[i];
var id = item.media$group.yt$videoid.$t;
var title = item.title.$t;
var seconds = item.media$group.yt$duration.seconds;
var media = new Media(id, title, seconds, "yt");
vids.push(media);
} catch(e) {
}
}
2013-10-11 21:31:40 +00:00
callback(false, vids);
2013-10-11 21:31:40 +00:00
var links = data.feed.link;
for(var i in links) {
if(links[i].rel === "next")
Getters["yp"](id, callback, links[i].href);
}
} catch(e) {
callback(e, null);
}
2013-10-11 21:31:40 +00:00
});
},
/* youtube.com search */
ytSearch: function (terms, callback) {
var sv = Server.getServer();
for(var i in terms)
terms[i] = encodeURIComponent(terms[i]);
var query = terms.join("+");
var options = {
host: "gdata.youtube.com",
port: 443,
path: "/feeds/api/videos/?q=" + query + "&v=2&alt=json",
method: "GET",
dataType: "jsonp",
timeout: 1000
};
2014-01-22 23:11:26 +00:00
if(Config.get("youtube-v2-key")) {
2013-10-11 21:31:40 +00:00
options.headers = {
2014-01-22 23:11:26 +00:00
"X-Gdata-Key": "key=" + Config.get("youtube-v2-key")
};
2013-10-11 21:31:40 +00:00
}
2013-10-11 21:31:40 +00:00
urlRetrieve(https, options, function (status, data) {
if(status !== 200) {
callback("YTSearch HTTP " + status, null);
return;
}
2013-10-11 21:31:40 +00:00
try {
data = JSON.parse(data);
var vids = [];
for(var i in data.feed.entry) {
try {
var item = data.feed.entry[i];
var id = item.media$group.yt$videoid.$t;
var title = item.title.$t;
var seconds = item.media$group.yt$duration.seconds;
var media = new Media(id, title, seconds, "yt");
media.thumb = item.media$group.media$thumbnail[0];
vids.push(media);
} catch(e) {
}
}
2013-10-11 21:31:40 +00:00
callback(false, vids);
} catch(e) {
callback(e, null);
}
});
},
/* vimeo.com */
vi: function (id, callback) {
var m = id.match(/([\w-]+)/);
if (m) {
id = m[1];
} else {
callback("Invalid ID", null);
return;
}
if (Config.get("vimeo-oauth.enabled")) {
return Getters.vi_oauth(id, callback);
}
2013-10-11 21:31:40 +00:00
var options = {
host: "vimeo.com",
port: 443,
path: "/api/v2/video/" + id + ".json",
method: "GET",
dataType: "jsonp",
timeout: 1000
};
urlRetrieve(https, options, function (status, data) {
if(status === 404) {
callback("Video not found", null);
return;
} else if(status === 403) {
callback("Private video", null);
return;
} else if(status === 503) {
callback("API failure", null);
return;
} else if(status !== 200) {
callback("YTv2 HTTP " + status, null);
return;
}
2013-08-01 03:26:11 +00:00
2013-10-11 21:31:40 +00:00
try {
data = JSON.parse(data);
data = data[0];
var seconds = data.duration;
var title = data.title;
var media = new Media(id, title, seconds, "vi");
callback(false, media);
} catch(e) {
var err = e;
if(buffer.match(/not found/))
err = "Video not found";
callback(err, null);
}
});
},
vi_oauth: function (id, callback) {
var OAuth = require("oauth");
var oa = new OAuth.OAuth(
"https://vimeo.com/oauth/request_token",
"https://vimeo.com/oauth/access_token",
Config.get("vimeo-oauth.consumer-key"),
Config.get("vimeo-oauth.secret"),
"1.0",
null,
"HMAC-SHA1"
);
oa.get("https://vimeo.com/api/rest/v2?format=json" +
"&method=vimeo.videos.getInfo&video_id=" + id,
null,
null,
function (err, data, res) {
if (err) {
return callback(err, null);
}
try {
data = JSON.parse(data);
if (data.stat !== "ok") {
return callback(data.err.msg, null);
}
var video = data.video[0];
if (video.embed_privacy !== "anywhere") {
return callback("Embedding disabled", null);
}
var id = video.id;
var seconds = parseInt(video.duration);
var title = video.title;
callback(null, new Media(id, title, seconds, "vi"));
} catch (e) {
callback("Error handling Vimeo response", null);
}
});
},
2013-10-11 21:31:40 +00:00
/* dailymotion.com */
dm: function (id, callback) {
// Dailymotion's API is an example of an API done right
// - Supports SSL
// - I can ask for exactly which fields I want
// - URL is simple
// - Field names are sensible
// Other media providers take notes, please
var m = id.match(/([\w-]+)/);
if (m) {
id = m[1];
} else {
callback("Invalid ID", null);
return;
}
var options = {
host: "api.dailymotion.com",
port: 443,
path: "/video/" + id + "?fields=duration,title",
method: "GET",
dataType: "jsonp",
timeout: 1000
};
urlRetrieve(https, options, function (status, data) {
if (status === 404) {
callback("Video not found", null);
return;
} else if (status !== 200) {
callback("DM HTTP " + status, null);
return;
}
2013-08-01 03:26:11 +00:00
2013-10-11 21:31:40 +00:00
try {
data = JSON.parse(data);
var title = data.title;
var seconds = data.duration;
if(title === "Deleted video" && seconds === 10) {
2013-10-07 05:10:16 +00:00
callback("Video not found", null);
return;
2013-08-01 03:26:11 +00:00
}
2013-10-11 21:31:40 +00:00
var media = new Media(id, title, seconds, "dm");
callback(false, media);
} catch(e) {
callback(err, null);
}
});
},
/* soundcloud.com */
sc: function (id, callback) {
// Soundcloud's API is badly designed and badly documented
// In order to lookup track data from a URL, I have to first
// make a call to /resolve to get the track id, then make a second
// call to /tracks/{track.id} to actally get useful data
// This is a waste of bandwidth and a pain in the ass
const SC_CLIENT = "2e0c82ab5a020f3a7509318146128abd";
var m = id.match(/([\w-\/\.:]+)/);
if (m) {
id = m[1];
} else {
callback("Invalid ID", null);
return;
}
2013-08-01 03:26:11 +00:00
2013-10-11 21:31:40 +00:00
var options = {
host: "api.soundcloud.com",
port: 443,
path: "/resolve.json?url=" + id + "&client_id=" + SC_CLIENT,
method: "GET",
dataType: "jsonp",
timeout: 1000
};
urlRetrieve(https, options, function (status, data) {
if(status === 404) {
callback("Sound not found", null);
return;
} else if(status === 503) {
callback("API failure", null);
return;
} else if(status !== 302) {
callback("SC HTTP " + status, null);
return;
}
2013-10-11 21:31:40 +00:00
var track = null;
try {
data = JSON.parse(data);
track = data.location;
} catch(e) {
callback(e, null);
return;
}
2013-10-11 21:31:40 +00:00
var options2 = {
host: "api.soundcloud.com",
port: 443,
2013-10-11 21:31:40 +00:00
path: track,
method: "GET",
dataType: "jsonp",
timeout: 1000
};
2013-10-11 21:31:40 +00:00
// I want to get off async's wild ride
urlRetrieve(https, options2, function (status, data) {
if(status !== 200) {
2013-10-07 05:10:16 +00:00
callback("SC HTTP " + status, null);
return;
}
try {
data = JSON.parse(data);
2013-10-11 21:31:40 +00:00
// Duration is in ms, but I want s
var seconds = data.duration / 1000;
var title = data.title;
var media = new Media(id, title, seconds, "sc");
callback(false, media);
} catch(e) {
2013-10-07 05:10:16 +00:00
callback(e, null);
}
});
2013-10-11 21:31:40 +00:00
});
},
/* livestream.com */
li: function (id, callback) {
var m = id.match(/([\w-]+)/);
if (m) {
id = m[1];
} else {
callback("Invalid ID", null);
return;
}
var title = "Livestream.com - " + id;
var media = new Media(id, title, "--:--", "li");
callback(false, media);
},
/* twitch.tv */
tw: function (id, callback) {
var m = id.match(/([\w-]+)/);
if (m) {
id = m[1];
} else {
callback("Invalid ID", null);
return;
}
var title = "Twitch.tv - " + id;
var media = new Media(id, title, "--:--", "tw");
callback(false, media);
},
/* justin.tv */
jt: function (id, callback) {
var m = id.match(/([\w-]+)/);
if (m) {
id = m[1];
} else {
callback("Invalid ID", null);
return;
}
var title = "Justin.tv - " + id;
var media = new Media(id, title, "--:--", "jt");
callback(false, media);
},
/* ustream.tv */
us: function (id, callback) {
// 2013-09-17
// They couldn't fucking decide whether channels should
// be at http://www.ustream.tv/channel/foo or just
// http://www.ustream.tv/foo so they do both.
// [](/cleese)
var m = id.match(/([^\?&#]+)|(channel\/[^\?&#]+)/);
if (m) {
id = m[1];
} else {
callback("Invalid ID", null);
return;
}
var options = {
host: "www.ustream.tv",
port: 80,
path: "/" + id,
method: "GET",
timeout: 1000
};
urlRetrieve(http, options, function (status, data) {
if(status !== 200) {
callback("Ustream HTTP " + status, null);
return;
}
2013-10-11 21:31:40 +00:00
// Regexing the ID out of the HTML because
// Ustream's API is so horribly documented
// I literally could not figure out how to retrieve
// this information.
//
// [](/eatadick)
var m = data.match(/cid":([0-9]+)/);
if(m) {
var title = "Ustream.tv - " + id;
var media = new Media(m[1], title, "--:--", "us");
callback(false, media);
} else {
2013-10-11 21:31:40 +00:00
callback(true, null);
}
2013-10-11 21:31:40 +00:00
});
},
/* JWPlayer */
jw: function (id, callback) {
var title = "JWPlayer - " + id;
var media = new Media(id, title, "--:--", "jw");
callback(false, media);
},
/* rtmp stream */
rt: function (id, callback) {
var title = "Livestream";
var media = new Media(id, title, "--:--", "rt");
callback(false, media);
},
/* imgur.com albums */
im: function (id, callback) {
var m = id.match(/([\w-]+)/);
if (m) {
id = m[1];
} else {
callback("Invalid ID", null);
return;
}
2013-10-11 21:31:40 +00:00
var title = "Imgur Album - " + id;
var media = new Media(id, title, "--:--", "im");
callback(false, media);
},
/* custom embed */
cu: function (id, callback) {
id = CustomEmbedFilter(id);
var media = new Media(id, "Custom Media", "--:--", "cu");
callback(false, media);
2013-11-07 23:19:36 +00:00
},
/* google docs */
gd: function (id, callback) {
var options = {
host: "docs.google.com",
path: "/file/d/" + id + "/edit",
port: 443
};
urlRetrieve(https, options, function (status, res) {
if (status !== 200) {
callback("Google Docs rejected: HTTP " + status, false);
return;
}
var m = res.match(/main\((.*?)\);<\/script>/);
if (m) {
try {
var data = m[1];
data = data.substring(data.indexOf(",") + 1);
data = data.replace(/'(.*?)'([:\,\}\]])/g, "\"$1\"$2");
data = data.replace(/\\x(\d*)/g, function (sub, s1) {
return String.fromCharCode(parseInt(s1, 16));
});
2013-11-07 23:19:36 +00:00
data = "[" + data + "]";
var js = JSON.parse(data);
var title = js[0].title;
var seconds = js[1].videodetails.duration / 1000;
var med = new Media(id, title, seconds, "gd");
2013-11-07 23:19:36 +00:00
var fv = js[1].videoplay.flashVars;
var fvstr = "";
for (var k in fv) {
if (k === "autoplay")
fv[k] = "1";
fvstr += "&" + k + "=" + encodeURIComponent(fv[k]);
}
fvstr = fvstr.substring(1);
var url = js[1].videoplay.swfUrl + "&enablejsapi=1";
med.object = {
type: "application/x-shockwave-flash",
allowscriptaccess: "always",
allowfullscreen: "true",
wmode: "opaque",
data: url
};
med.params = [
{
name: "allowFullScreen",
value: "true"
},
{
name: "allowScriptAccess",
value: "always"
},
{
name: "wmode",
value: "opaque"
},
{
name: "flashvars",
value: fvstr
}
];
callback(false, med);
} catch (e) {
callback("Parsing of Google Docs output failed", null);
}
} else {
callback(res, null);
}
});
2013-10-11 21:31:40 +00:00
}
};
/**
* Function to workaround Vimeo being a dick and blocking my domain from embeds.
* Retrieves the player page and extracts the direct links to the MP4 encoded videos.
*/
function vimeoWorkaround(id, cb) {
if (typeof cb !== "function") {
return;
}
var failcount = 0;
var inner = function () {
var options = {
host: "player.vimeo.com",
path: "/video/" + id,
headers: {
"User-Agent": "Mozilla/5.0 (X11; Linux x86_64; rv:29.0) Gecko/20100101 Firefox/29.0",
"Referrer": "player.vimeo.com"
}
};
var parse = function (data) {
2014-02-05 04:02:27 +00:00
var i = data.indexOf("{\"cdn_url\"");
if (i === -1) {
Logger.errlog.log("Vimeo workaround failed (i=-1): http://vimeo.com/" + id);
setImmediate(function () {
cb({});
});
return;
}
var j = data.indexOf("};", i);
2014-02-05 04:02:27 +00:00
var json = data.substring(i, j+1);
try {
json = JSON.parse(json);
var codec = json.request.files.codecs[0];
var files = json.request.files[codec];
setImmediate(function () {
cb(files);
});
} catch (e) {
// This shouldn't happen due to the user-agent, but just in case
if (data.indexOf("crawler") !== -1) {
Logger.syslog.log("Warning: VimeoIsADoucheCopter got crawler response");
failcount++;
if (failcount > 4) {
Logger.errlog.log("VimeoIsADoucheCopter got bad response 5 times!"+
" Giving up.");
setImmediate(function () {
cb({});
});
} else {
setImmediate(function () {
inner();
});
}
return;
} else if (data.indexOf("This video does not exist.") !== -1) {
cb({});
return;
} else if (data.indexOf("Because of its privacy settings, this video cannot be played here.") !== -1) {
cb({});
}
Logger.errlog.log("Vimeo workaround error: ");
Logger.errlog.log(e);
2014-02-05 04:02:27 +00:00
Logger.errlog.log("http://vimeo.com/" + id);
setImmediate(function () {
cb({});
});
}
};
http.get(options, function (res) {
res.setEncoding("utf-8");
var buffer = "";
res.on("data", function (data) {
buffer += data;
});
res.on("end", function () {
parse(buffer);
});
});
};
inner();
}
2013-10-11 21:31:40 +00:00
module.exports = {
Getters: Getters,
getMedia: function (id, type, callback) {
if(type in this.Getters) {
this.Getters[type](id, callback);
} else {
callback("Unknown media type '" + type + "'", null);
}
},
vimeoWorkaround: vimeoWorkaround
2013-10-11 21:31:40 +00:00
};