To install wget in different Linux flavours, try as follows:
- For Debian, Ubuntu via apt-get :
# apt-get install wget
- For Fedora, CentOS via yum :
# yum -y install wget
Cipher API ver 1.1 update and MD5 hash
After this update, API includes one extra param in its response which is the MD5 hash of the HTML5 player file content.
This has been employed to track if the algorithm changes for the current HTML5 player. If it changed, then it will update the algo dictionary.
Even though API has its filter to check the file changes, this is also the responsibility of the client to check the status of the HTML5 player. This is even more critical when the client saved the dictionary in their local environment for faster processing, which is also the expected method to access the API.
MD5 hash of the file can be checked as follows:
/*
* MD5 hash calculation of the content of the html5 player id en_US-vflr38Js6
*/
md5_file('http://s.ytimg.com/yts/jsbin/html5player-en_US-vflr38Js6/html5player.js');
First check the MD5 hash and match it with the algo dictionary MD5 hash. If both matches then file not changed otherwise make a new API call for the updated dictionary.
API is itself capable of checking the changes if any. Just you have to be aware of the file changes.
Demo example based on Cipher API
Cipher API provides the algorithm dictionary which can be used for decrypting cipher codes.
Below is the demo example for the same which is also released in Github at https://github.com/bitnol/CipherAPI
<?php
/**
* Author: Akhilesh Chandra Gupta
* Author URI: http://www.bitnol.com
* API URI: http://api.gitnol.com
* Description: Demo of the application of Cipher API
* License: MIT License
*/
// Video with cipher signature
$video_id = "zDrNLZ1uJ2w";
// get_video_info url formation
// although for cipher signature we have to get the details from the video's webapge not from get_video_info object
$info_url = "http://www.youtube.com/get_video_info?el=detailpage&asv=3&video_id=".$video_id;
// youtube webpage url formation
$yt_url = 'http://www.youtube.com/watch?v='.$video_id;
// get the contents from the url
$raw_data = file_get_contents($info_url);
// parse the data received and save it as an array
$output = array();
parse_str($raw_data,$output);
// check the status of the get_video_info object
if($output['status']=='ok'){
// check for the cipher signature
$cipher = (isset($output['use_cipher_signature']) && $output['use_cipher_signature']=='True') ? true : false;
// If cipher is true then we have to decode it
if($cipher == true){
// if cipher is true then we have to change the plan and get the details from the video's youtube wbe page
$yt_html = file_get_contents($yt_url);
// parse for the script containing the configuration
preg_match('/ytplayer.config = {(.*?)};/',$yt_html,$match);
$yt_object = @json_decode('{'.$match[1].'}') ;
/// check if we are able to parse data
if(!is_object($yt_object)){
echo 'Sorry! Unable to parse Data';
}else{
// parse available formats
$formats = $yt_object->args->url_encoded_fmt_stream_map;
// get the player id from assets section
$player_id = strbtwn($yt_object->assets->js,'html5player-','.js');
$player_id = explode("/", $player_id);
$player_id = $player_id[0];
echo 'Player ID: '.$player_id.'<br /><hr />';
// get the algo dictionary
// first check if the file exists
if(file_exists('./algo.json'))
$algos = json_decode(file_get_contents('algo.json'),true);
else{
// API call to fetch the algo dictionary
$algos_dict = file_get_contents("http://api.gitnol.com/getAlgo.php?playerID=".$player_id);
// saving the algo dictonary in local env for easy access
// Note: Developers should save the dictionary in their local env.
// Only make the API call for the new player ids which is not present in the algo dictionary.
// Repeated violation will results in IP ban.
file_put_contents('algo.json', $algos_dict);
$algos = json_decode($algos_dict,true);
}
/// check if the algo exist for the given player id
if(!array_key_exists($player_id, $algos)){
// if the algo dictionary is old then fetch a new one
$algos_dict = file_get_contents("http://api.gitnol.com/getAlgo.php?playerID=".$player_id);
file_put_contents('algo.json', $algos_dict);
$algos = json_decode($algos_dict,true);
$algo = $algos[$player_id][1];
}else{
$algo = $algos[$player_id][1];
}
echo 'Algo Used: '.$algo.'<br /><hr />';
// download links formation
$dlinks = array();
$links = explode(',',$formats);
echo 'Download links <br /><br />';
foreach ($links as $link) {
parse_str($link,$linkarr);
// parse link array one by one and decrypt the signature
$dlinks[$linkarr['itag']] = $linkarr['url'] . "&signature=" . decrypt($linkarr['s'],$algo);
echo $linkarr['itag'].'<br />';
echo $dlinks[$linkarr['itag']].'<br /><br />';
}
echo '<hr />';
}
}else{
echo 'Video Is not cipher not needed to decode';
}
}else{
echo 'Unable to get Video Info';
}
// string helper function
function strbtwn($content,$start,$end){
$r = explode($start, $content);
if (isset($r[1])){
$r = explode($end, $r[1]);
return $r[0];
}
return '';
}
// signature decoding
// parse the python string operation into php string operation
function decrypt($sig, $algo){
$funcarr = explode(' + ', $algo);
$decrypt = '';
foreach($funcarr as $singfunc){
$singfunc = substr($singfunc,2,-1);
$operators = explode(':', $singfunc);
if (sizeof($operators) == 1) {
$decrypt .= $sig[$operators[0]];
}
if (sizeof($operators) == 2) {
if($operators[0] == ''){
$decrypt .= substr($sig, 0 ,$operators[1]);
}
if($operators[1] == ''){
$decrypt .= substr($sig, $operators[0]);
}
if($operators[0] >= 0 && $operators[1] >= 0){
$decrypt .= substr($sig, $operators[0], $operators[1] - $operators[0]);
}
}
if (sizeof($operators) == 3) {
if($operators[0] == '' && $operators[1] == ''){
$decrypt .= strrev($sig);
}
if($operators[0] >=0 && $operators[1] == '' && $operators[0] != ''){
$decrypt .= strrev(substr($sig, 0, $operators[0] + 1));
}
if($operators[0] >=0 && $operators[1] >= 0 && $operators[0] != '' && $operators[1] != ''){
$decrypt .= strrev(substr($sig, $operators[1] + 1, $operators[0] - $operators[1]));
}
}
}
return $decrypt;
}
?>You can also fork it from Github. Please do leave feedback if any.
A note of Cipher Signature on Stackoverflow
How to use Cipher API to decrypt signatures
Before proceeding please check the home page of Cipher API: http://api.gitnol.com
We can retrieve the signature decrypting algorithm using two method:
- By passing just player ID i.e. http://api.gitnol.com/getAlgo.php?playerID=en_US-vflz7mN60
- By passing player ID and signature format i.e. http://api.gitnol.com/getAlgo.php?playerID=en_US-vflz7mN60&sigformat=42.40
Developers can use any of them depending upon their requirement. But I recommend to use first method because first method includes the decrypting algorithm of most expected signature format.
The decrypting algorithm is in Python syntax. So developers working with PHP can use the following function to decrypt the signature using the algorithm obtained from first method:
function decrypt($sig, $algo){
$funcarr = explode(' + ', $algo);
$decrypt = '';
foreach($funcarr as $singfunc){
$singfunc = substr($singfunc,2,-1);
$operators = explode(':', $singfunc);
if (sizeof($operators) == 1) {
$decrypt .= $sig[$operators[0]];
}
if (sizeof($operators) == 2) {
if($operators[0] == ''){
$decrypt .= substr($sig, 0 ,$operators[1]);
}
if($operators[1] == ''){
$decrypt .= substr($sig, $operators[0]);
}
if($operators[0] >= 0 && $operators[1] >= 0){
$decrypt .= substr($sig, $operators[0], $operators[1] - $operators[0]);
}
}
if (sizeof($operators) == 3) {
if($operators[0] == '' && $operators[1] == ''){
$decrypt .= strrev($sig);
}
if($operators[0] >=0 && $operators[1] == '' && $operators[0] != ''){
$decrypt .= strrev(substr($sig, 0, $operators[0] + 1));
}
if($operators[0] >=0 && $operators[1] >= 0 && $operators[0] != '' && $operators[1] != ''){
$decrypt .= strrev(substr($sig, $operators[1] + 1, $operators[0] - $operators[1]));
}
}
}
return $decrypt;
}Use above function to with proper params i.e. decrypt($signature, $algorithm) and it will return the decrypted signature.
Hopefully I have considered each case of Python syntax. If anybody found any missing case then please contribute it.
More articles will follow soon on this section.