Need help with finding stream url

[paul92]

Hi all

I'm looking for a way to pull the streams from the broadcaster in my country (SABC) to a m3u format, so that I can connect them into my PVR.

They seem to be in mp4 segments. I have attached a screenshot detailing one channel
[Attachment 68267 - Click to enlarge]

There's probably about 4 channels needing to be pulled to a m3u format and streamed. If the steaming token expires, I would need something to renew the token automatically.

Interlark has designed something similar for another service https://github.com/interlark/ustvgo-iptv

Assuming I would need something like this, unfortunately I have no idea how to use Python, or where to even start, but I love to learn.

So if anyone can offer me detailed info, on how to do this please let me know.

If I need to submit more info, please let me know what.

Thanks in advance. (:

[jack_666]

Hi paul92, you can help your cause greatly if you provide URL of the streams that you desire.

If login is required then any action for your cause will be greatly demished.


If your provider, SABC, uses encryption then you may be SOL as these require keys and these keys usually expire.

unfortunately I have no idea how to use Python, or where to even start, but I love to learn

Your lack of knowledge on programming is excusable but "not knowing where to start" is not. Ever heard of YouTube?, what about books?

[jack_666]

Ok, I did some checking on my own

https://www.sabcplus.com/en/account/signup

Need mobile number to access.


I know of fake email addresses but fake mobile numbers with sms eludes me.

[Quint]

Try this:
https://receive-sms.cc/France-Phone-Number/33644623966
France and European numbers worked for me where others didn't.

[paul92]

Ok, I did some checking on my own

https://www.sabcplus.com/en/account/signup

Need mobile number to access.


I know of fake email addresses but fake mobile numbers with sms eludes me.

Check your PMs.

I sent my registered login details

Paul

[paul92]

it's been quite a while since I last even thought about this - but had a look at this again, this time my knowledge is slightly better. First thing is that the channels are geoblocked to South African IP address, so one would need a South African VPN. Secondly from what I've found is that there seems to be some sort of token protection, as a unique token is provided, upon each connection, which seems to be valid for at least close to 4 hours. Using an extension called The Stream detector, I was able to grab the stream URL (which eventually expire) pasted it into VLC, and it played immediately - which would seem to me that the only protection there, is that there needs to be a tokenID.

[2nHxWW6GkN1l916N3ayz8HQoi]

Out of curiosity, what's the site and could you give a few stream URLs? The link @jack wrote is dead

[paul92]

Out of curiosity, what's the site and could you give a few stream URLs? The link @jack wrote is dead

website is

https://sabc-plus.com/

requires registration - but is free to register.

URLs to streams
SABC1

https://sabc-plus.com/live/320/SABC-1

SABC2

https://sabc-plus.com/live/321/SABC-2

SABC3

https://sabc-plus.com/live/322/S3

SABC Sport

https://sabc-plus.com/live/323/SABC-Sport

SABC Education

https://sabc-plus.com/live/325/SABC-Education

SABC News

https://sabc-plus.com/live/324/SABC-News

SABC Lehae

https://sabc-plus.com/live/326/SABC-Lehae

Using The Stream Detector, I was able to get the following URLS

SABC1

 https://sabconeta.cdn.mangomolo.com/sabc1/smil:sabc1.stream.smil/chunklist_b1600000_t64NzIwcA==.m3u8

SABC2

 https://sabctwota.cdn.mangomolo.com/sabc2/smil:sabc2.stream.smil/chunklist_b1600000_t64NzIwcA==.m3u8

SABC3

 https://sabctreta.cdn.mangomolo.com/sabc3/smil:sabc3.stream.smil/chunklist_b1600000_t64NzIwcA==.m3u8

SABC Sport

 https://sabctwotasp.cdn.mangomolo.com/sport/smil:sport.stream.smil/chunklist_b1600000_t64NzIwcA==.m3u8

SABC Education

 https://sabctretaed.cdn.mangomolo.com/edu/smil:edu.stream.smil/chunklist_b2000000_t64NzIwcA==.m3u8

SABC News

 https://sabconetanw.cdn.mangomolo.com/news/smil:news.stream.smil/chunklist_b250000_t64MjQwcA==.m3u8

SABC Lehae

 https://sabctretalh.cdn.mangomolo.com/lehae/smil:lehae.stream.smil/chunklist_b2000000_t64NzIwcA==.m3u8

these open, and play fine with VLC, but was advised they will probably eventually expire




so need a way to in a way ) at least from my understanding
Automatically direct

https://sabc-plus.com/live/320/SABC-1

to fetch this URL [/code]

https://sabconeta.cdn.mangomolo.com/sabc1/smil:sabc1.stream.smil/chunklist_b1600000_t64NzIwcA==.m3u8

I have noted in earlier experimentation that this is based on token based authentication, and that it seems to come from something in the cookies file. Happy to provide further info if needed. Just a reminder these will only open with a South African IP address - I'm currently in New Zealand, and using my VPN to do all this.

[StreamBabyStream]

I wrote a forwarder for these channels a while ago (except SABC Education and SABC Lehae that I missed before they enforced registration). It uses the embedded players (hosted at Mangomolo) and extracts a tokenized m3u8 url from it. I just tested it real quick and as long as both the player domain and the m3u8 domain are called with a South African ip, things still work fine for those channels.

Someone else is going to have to write the code as I'm really bad with Python, but here's how to do it:

Open this url, filling in two fixed id's (this is the player page):

https://player.mangomolo.com/v1/live?id=MTk0&channelid=<channelid>&countries=Q0M=&filter=DENY&signature=<signature>

channelid and signature for the channels that I know of are listed in the code below.

The resulting html pages will have the tokenized m3u8 link in them, in a long block of Javascript. Really easy to extract with e.g. this regular expression (this is a Python regular expression):

r"src: \"(.+)\""

And then you can just forward to this url. This is the code I use in Node.js (just the SABC part, it's using Express.js as the server framework to link requests and responses)

import got from 'got'

export default async function (req, res) {
  const channelParameters = {
    news: {
      channelid: 'MzI0',
      signature: '54e386ff3a9b282f709bdbe3cb7f026f'
    },
    1: {
      channelid: 'MzIw',
      signature: 'afef9c2e6bf161cd3b3d929cfa9d51cb'
    },
    2: {
      channelid: 'MzIx',
      signature: '9d92dff59c1eddf29e8f8d87ee3e78da'
    },
    3: {
      channelid: 'MzIy',
      signature: 'c8cb953e450cc1b2b851663b64928cfe'
    },
    sport: {
      channelid: 'MzIz',
      signature: 'bce106f2fb94504e2a3d19cd0c0175cc'
    }
  }

  const { id } = req.params

  // Get url from player page
  const { channelid, signature } = channelParameters[id]

  const playerPage = await got(`https://player.mangomolo.com/v1/live?id=MTk0&channelid=${channelid}&countries=Q0M=&filter=DENY&signature=${signature}`)
  const url = /src: "(.+)"/g.exec(playerPage.body)[1]

  res.redirect(url)
}

[2nHxWW6GkN1l916N3ayz8HQoi]

Happy to provide further info if needed.

Seems like an easy site to write a script for. Unfortunately, they require phone numbers for registration and all virtual ones are toasted. Without access to an account, I can't do much. Try the solution mentioned above.

[imr_saleh]

I wrote a forwarder for these channels a while ago (except SABC Education and SABC Lehae that I missed before they enforced registration). It uses the embedded players (hosted at Mangomolo) and extracts a tokenized m3u8 url from it. I just tested it real quick and as long as both the player domain and the m3u8 domain are called with a South African ip, things still work fine for those channels.

Someone else is going to have to write the code as I'm really bad with Python, but here's how to do it:

Open this url, filling in two fixed id's (this is the player page):

https://player.mangomolo.com/v1/live?id=MTk0&channelid=<channelid>&countries=Q0M=&filter=DENY&signature=<signature>

channelid and signature for the channels that I know of are listed in the code below.

The resulting html pages will have the tokenized m3u8 link in them, in a long block of Javascript. Really easy to extract with e.g. this regular expression (this is a Python regular expression):

r"src: \"(.+)\""

And then you can just forward to this url. This is the code I use in Node.js (just the SABC part, it's using Express.js as the server framework to link requests and responses)

import got from 'got'

export default async function (req, res) {
  const channelParameters = {
    news: {
      channelid: 'MzI0',
      signature: '54e386ff3a9b282f709bdbe3cb7f026f'
    },
    1: {
      channelid: 'MzIw',
      signature: 'afef9c2e6bf161cd3b3d929cfa9d51cb'
    },
    2: {
      channelid: 'MzIx',
      signature: '9d92dff59c1eddf29e8f8d87ee3e78da'
    },
    3: {
      channelid: 'MzIy',
      signature: 'c8cb953e450cc1b2b851663b64928cfe'
    },
    sport: {
      channelid: 'MzIz',
      signature: 'bce106f2fb94504e2a3d19cd0c0175cc'
    }
  }

  const { id } = req.params

  // Get url from player page
  const { channelid, signature } = channelParameters[id]

  const playerPage = await got(`https://player.mangomolo.com/v1/live?id=MTk0&channelid=${channelid}&countries=Q0M=&filter=DENY&signature=${signature}`)
  const url = /src: "(.+)"/g.exec(playerPage.body)[1]

  res.redirect(url)
}

news  http://127.0.0.1:3000/MzI0/54e386ff3a9b282f709bdbe3cb7f026f/hls.m3u8
1  http://127.0.0.1:3000/MzIw/afef9c2e6bf161cd3b3d929cfa9d51cb/hls.m3u8
2  http://127.0.0.1:3000/MzIx/9d92dff59c1eddf29e8f8d87ee3e78da/hls.m3u8
3  http://127.0.0.1:3000/MzIy/c8cb953e450cc1b2b851663b64928cfe/hls.m3u8
sport  http://127.0.0.1:3000/MzIz/bce106f2fb94504e2a3d19cd0c0175cc/hls.m3u8

from flask import Flask, redirect, Response
import re
import requests

app = Flask(__name__)

@app.route('/<path:url>', methods=['GET'])
def on_fetch(url):
    parts = url.split('/')
    if len(parts) < 3:
        return Response('Invalid URL', status=400)

    channelid = parts[-3]
    signature = parts[-2]
    api_response = requests.get(f'https://player.mangomolo.com/v1/live?id=MTk0&channelid={channelid}&countries=Q0M=&filter=DENY&signature={signature}')

    if api_response.status_code != 200:
        return Response('geo blocked', status=403)

    match = re.search(r'src: "(.*)"', api_response.text)
    if match:
        m3u8 = match.group(1)
        return redirect(m3u8, code=301)
    else:
        return Response('m3u8 URL not found', status=404)

if __name__ == '__main__':
    app.run(port=3000)

[paul92]

Thanks everyone - was able to get this up and running. No buffering, and works brilliantly.

Next thing is how can I convert the EPG to an epg_xml file or whatever, so I can add the EPG as a guide source.

Thanks in advance.

[StreamBabyStream]

Thanks imr_saleh! Also: I should make it a habit some day to check for and report errors in this kind of code

paul92: where would you get EPG info for these channels? From the SABC site itself? I'm currently writing some scripts to get EPG data from a few channels that I watch on a regular basis and always interested to add more.

[paul92]

Thanks imr_saleh! Also: I should make it a habit some day to check for and report errors in this kind of code

paul92: where would you get EPG info for these channels? From the SABC site itself? I'm currently writing some scripts to get EPG data from a few channels that I watch on a regular basis and always interested to add more.

Should be easy enough get make a XMLTV script to pull from the SABC-plus website, if going by how easy it was to get the live channels working after I figured out what to do. Even Education works now. I asked ChatGPT to make a flask script, installed it as a service on my JHB VPS, pointed NextPVR installed on JHB VPS to those links (to deal with geo-blocking) then forwarded that nextpvr m3u8 to my main PVR VPS. (I have a netmaker WAN which is essentially a high speed Wireguard mesh network)

and likewise. Not much of a TV person, but there a few channels I watch, but also enjoy the technology side, and building things, and makings work. I use Tinuva has a good EPG which covers a lot of DStv channels, but so much SABC.

[paul92]

Thanks imr_saleh! Also: I should make it a habit some day to check for and report errors in this kind of code

paul92: where would you get EPG info for these channels? From the SABC site itself? I'm currently writing some scripts to get EPG data from a few channels that I watch on a regular basis and always interested to add more.

Hi any idea what's changed - this was working beautifully but when I checked again - the streams were failing to load

here's my script

from flask import Flask, redirect, abort
import requests
import re

app = Flask(__name__)

# Define the channel parameters
channel_parameters = {
    'news': {
        'channelid': 'MzI0',
        'signature': '54e386ff3a9b282f709bdbe3cb7f026f'
    },
    '1': {
        'channelid': 'MzIw',
        'signature': 'afef9c2e6bf161cd3b3d929cfa9d51cb'
    },
    '2': {
        'channelid': 'MzIx',
        'signature': '9d92dff59c1eddf29e8f8d87ee3e78da'
    },
    '3': {
        'channelid': 'MzIy',
        'signature': 'c8cb953e450cc1b2b851663b64928cfe'
    },
    'sport': {
        'channelid': 'MzIz',
        'signature': 'bce106f2fb94504e2a3d19cd0c0175cc'
    },
    'education': {
        'channelid': 'MzI5',
        'signature': 'b7b9cef40ca3743a4bb32e528da2df67'
    },
    'lehae': { 
        'channelid': 'MzI2',
        'signature': '414ba5b272e6c9bdfc96faafa454406a'
    }
}

@app.route('/stream/<channel_id>')
def stream(channel_id):
    params = channel_parameters.get(channel_id)
    if not params:
        abort(404)  # Channel not found

    channelid = params['channelid']
    signature = params['signature']

    # Get URL from player page
    url = f"https://player.mangomolo.com/v1/live?id=MTk0&channelid={channelid}&countries=Q0M=&filter=DENY&signature={signature}"
    response = requests.get(url)

    # Extract the streaming URL from the response
    match = re.search(r'src: "(.+)"', response.text)
    if match:
        stream_url = match.group(1)
        return redirect(stream_url)
    else:
        abort(500)  # Error extracting URL

if __name__ == '__main__':
    app.run(debug=True)

when curling SABC3 I get this

paul@paul-Lenovo-ideapad-510-15ISK:~$ curl http://127.0.0.1:5000/stream/3
<!doctype html>
<html lang=en>
<title>500 Internal Server Error</title>
<h1>Internal Server Error</h1>
<p>The server encountered an internal error and was unable to complete your request. Either the server is overloaded or there is an error in the application.</p>

some URL has obviously been changed somewhere...

Hopefully someone here will be able to figure out just what changed exactly. I would have a look, but my brain isn't in code mode at the moment....

Thanks

[imr_saleh]

They made some changes, need access

[StreamBabyStream]

Well, now I'm paranoid. I'd been using this method with these exact same parameters for the last 8 months, always worked just fine. Now, just a short while after sharing the id's and signatures things have been changed around.

So while I found a pretty easy solution, I'm a bit hesitant to share it here. Because it's probably just as easy to patch. On the other hand, I'm pretty new here, so I'm not sure if being 'secretive' and only sharing things by PM is appreciated on this forum.

[2nHxWW6GkN1l916N3ayz8HQoi]

On the other hand, I'm pretty new here, so I'm not sure if being 'secretive' and only sharing things by PM is appreciated on this forum.

There are some sites that are being monitored for tricks like yours. You're the one to decide but no one will judge you if you keep it to yourself or only share it to trusted forum members.

[StreamBabyStream]

There are some sites that are being monitored for tricks like yours. You're the one to decide but no one will judge you if you keep it to yourself or only share it to trusted forum members.

Got you, thanks! Seems silly to keep it purely to myself. But equally silly to just spell it out here.

[paul92]

There are some sites that are being monitored for tricks like yours. You're the one to decide but no one will judge you if you keep it to yourself or only share it to trusted forum members.

Got you, thanks! Seems silly to keep it purely to myself. But equally silly to just spell it out here.

Just sent you a PM

[paul92]

Response Content: This channel is not allowed on the current domain