twitchbot

automatic twitchdownloader with optional twitter-bot function

Help with Repo

⚠️ Pull Request Guidelines ⚠️

Pull requests should be done on the development branch only.

Tasks:

• Implement new features and improve existing ones to be better and more reliable.
• Restructure the program manly the twitterbot part(package it in the right way)

Guide

requirements

• (mini)conda
• requirements.txt
• a storage for the streams and yt credentials
• .env files for twitter/twitch/ifttt credential
• Linux/MacOS/windows(might requier a bit more tinkering in comparison to the first 2)

installation

1. clone the repo git clone https://github.com/iampickle/twitchbot.git

2. conda env

   conda create -n -tbot
   conda install pip

3. install requirements.txt pip install -r requirements.txt

4. customize channels.json -example-

    {
     "compress-server": {
       "ip": "192.168.188.12",
       "port": "8766"
     },
     "streamers": {
       "papaplatte": {
         "ytupload": true, //falg to enable yt-upload
         "tbot": {
           "start": 1, //start/end is for defining the 
           "end": 0.5, //puffer around the said word <puffer>|<word>|<puffer>
           "words": [ //if tbot obect exists also word-arry must be there
             "geil",
             "perfekt"
           ],
           "tiktokupload": true, //toggel tiktok upload cookies file need to be set in .env (https://github.com/wkaisertexas/tiktok-uploader#-authentication)
           "NOKEEP": true // if True folder of stream is deleted after Twitter-Bot is done
         }
       },
       "rezo": {} //channel without tbot object simply just record streams
     }
   }

5. create .env

   #important dirs
   program-dir=<path of program location (git folder)> #important for start.sh when used in systemd
   conda-path=/home/<user>/miniconda3/bin/activate # <= very important
   dir=</your/path>
   channel-config=<./channels.json>
   
   #rendering options e.g. "libx264"
   codec=<codec>
   
   #db
   db-host=<url/ip to database>
   db-user=<username>
   db-password=<password>
   db-database=<database name>
   
   #vosk
   vosk-model=<vosk-model-small-de-zamia-0.3>
   
   #twitter
   bearertoken=<your-bearertoken>
   consumer-key=<your-consumer-key>
   consumer-secret=<your-consumer-secret>
   access-token=<your-access-token>
   access-token-secret=<your-access-token-secret>
   
   #twitch
   Client-ID-Twitch=<your-Client-ID-Twitch>
   Authorization-Twitch=<your-Twitch-Authorization>
   
   #IFTTT
   Authorization-IFTTT=<your-IFTTT-Authorization>
   event=<your-event-name>

6. create dir for tbot example: /your/folder/...

   .
    └── mainfolder/ 
       ├── streamer1/ 
       │   └── steamer1-date/ 
       │       └── stream.mp4 
       ├── streamer2/
       │   └── steamer2-date/ 
       │       ├── analytics/
       │       │   ├── 2b762e54-a315-495c-bd62-f65ee4f5390e.png(viewer chart)
       │       │   ├── 2b762e54-a315-495c-bd62-f65ee4f5390e.png(word count)
       │       │   └── 2b762e54-a315-495c-bd62-f65ee4f5390e.png(top table)
       │       ├── output/
       │       │   └── stitched.mp4 
       │       └── stream.mp4 
       └── .yt-credentials/ 
           └── client_secret.json
   

7. .yt-credentials/ client_secret.json(tutorial) Verification of endpoint needs to be started(endless uncompletion)

   

Tools (usage info)

createcredential.py

-can be used if all requirements are rightfully installed

-for non local applications(server) script can be executed localy or ssh-tunnel can be used in order to login via google-account

start-twitterbot.py

-script needs to be copied into main-dir

-vars in function args need to be manipulated

License

MIT