Openload: extract download URL using PhantomJS

openload.js

// Usage: phantomjs openload.js <video_url>
// if that doesn't work try: phantomjs --ssl-protocol=any openload.js <video_url>

var separator = ' | ';
var page = require('webpage').create(),
  system = require('system'),
  id, match;

if(system.args.length < 2) {
  console.error('No URL provided');
  phantom.exit(1);
}
match = system.args[1].match(
  /https?:\/\/(?:openload\.(?:co|io)|oload\.tv)\/(?:f|embed)\/([\w\-]+)/);
if(match === null) {
  console.error('Could not find video ID in provided URL');
  phantom.exit(2);
}
id = match[1];

// thanks @Mello-Yello :)
page.onInitialized = function() {
  page.evaluate(function() {
    delete window._phantom;
    delete window.callPhantom;
  });
};
page.settings.userAgent = "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/56.0.2924.87 Safari/537.36";
page.open('https://openload.co/embed/' + id + '/', function(status) {
  var info = page.evaluate(function() {
    return {
      decoded_id: document.getElementById('streamuri').innerHTML,
      title: document.querySelector('meta[name="og:title"],'
        + 'meta[name=description]').content
    };
  });
  var url = 'https://openload.co/stream/' + info.decoded_id + '?mime=true';
  console.log(url + separator + info.title);
  phantom.exit();
});

I had to call phantomjs with --ssl-protocol=any otherwise page.open would fail.

Thanks, I forgot to add it here.

It's already added in this PR. :D

Does this have method restricted by API downloads? I plan to use this to mount Openload streams into JWPlayer.

It's the same URL as when you watch something on Openload in your browser. That means it won't work for different IP. It will probably expire too but don't know after how long. I was downloading something for 2-3h and it worked fine.

Just a little question, when a client goes to my website and does "phantomjs openload.js <video_url>" via exec(), the ip shown on openload url is my machine's IP, not my visitor/client IP. How can I fix this? Do you have any idea? Cheers.

You've got two options:

• stream the video through your server; downside: huge traffic on your server,
• run the code on clients side; downside: client has to download phantomjs and run the code for themselves.

Hi,
Thanks for this.
I will like to ask a few questions,
I added this line to my edited version of the code page.settings.resourceTimeout = 3000; hoping that should in case the page takes time to load it should timeout.
I use a Python subprocess to execute the JS. it works as expected but in some cases when I call the communicate() function the application just freezes and have to start all over.
Is there a better approach you will advise to handle this?
Thanks a lot.

This snippet works fine for me:

import subprocess

try:
   my_input = raw_input
except NameError:
   my_input = input

url = my_input('Enter URL: ')

p = subprocess.Popen(['phantomjs', '--ssl-protocol=any', 'openload.js',
    url], stdout=subprocess.PIPE, stderr=subprocess.PIPE)
out, err = p.communicate()
print(out)

Both in Python 2 and 3.

This is not working again. You must change streamuri with streamurj.

See, when I update it they will change it again. So I'm not gonna bother.

It seems like it does not work anymore even after running
sed -e "32 s/getElementById..streamur./querySelector('*[id^=streamur]/"
on this openload.js file.

I'm of course getting a similar link to "https://openload.co/stream/miEoI5oT8JE~1516281189~184.73.0.0~ImWW-Wbq?mime=true | bunny HTTP/1.1" but GETting this link only gets me a "HTTP/1.1 400 Bad Request" type response.

Is this only me or they somehow changed something else somewhere.

@gonubana just tested and
curl -L -k -o <filename> <generated URL>
works perfectly fine.

Just read 4 comments above before saying "it doesn't work".

I don't think it is working anymore
`
TypeError: null is not an object (evaluating 'document.getElementById('streamuri').innerHTML')

undefined:3
:7
`

Not work anymore :(

#streamuri and #streamurl have been empty for a couple of weeks now.

Did anyone find a solution for this?

This is the current magic:
console.log('https://openload.co/stream/' + page.content.match(/\w+\d+\d+.\d+.\d+.\d+~\w+/)[0] + '?mime=true');

it should be
console.log('https://oload.stream/stream/' + page.content.match(/\w+~\d+~\d+.\d+.\d+.\d+~\w+/)[0] + '?mime=true');
for more clear

Since there is a domain oload.download, regex should be
/https?:\/\/(?:openload\.(?:co|io)|oload\.tv|oload\.download)\/(?:f|embed)\/([\w\-]+)/)

And, i was looking into the deobfuscated jsnice code of the source and i saw this line

if ("toString" in sin && sin.toString().indexOf("[native code") != -1 && document.getElementById.toString().indexOf("[native code") == -1 || window.callPhantom || /Phantom/.test(navigator.userAgent) || window.__phantomas || next() || window.domAutomation || window.webdriver || document.documentElement.getAttribute("webdriver"))

I guess there is some anti-measure phantomjs in the site, but aparently its not working, because @gonubana comment does captures the download URL...

reduce regex complexity
console.log('https://oload.stream/stream/' + page.content.match(/\w+~\d+~[\d\.]+~\w+/)[0] + '?mime=true');