TS3AudioBot

master	develop

About

This is our open-source TeamSpeak 3 audiobot project since we haven't found any other open-source one so far.
The bot has come a long way is pretty stable by now, though somtimes he hangs up or needs some other maintenance.
For now I'd only recomment this bot on small servers since it doesn't cover any more complex right systems and relies on discipline.

How our Bot works

The TS3AudioBot connects with at least 1 TeamSpeak3 Client instance wich allows you to:

issue commands to that instance.
play music for your channel.
tell him to stream to different Channels and/or Users simultaneously with TeamSpeak's whisper feature.

We use a self written TeamSpeak3 Client which gives us very low memory and cpu usage.
About 35MB Ram with 1500+ songs in history indexed
And 3-7% CPU usage on a single shared vCore from a Intel Xeon E5-1650 v2 @ 3.50GHz

Features & Plannings

Done:

Extract Youtube and Soundcloud songs as well as stream Twitch
Extensive history manager, including features like:
- getting the last x played songs
- get last x songs played by a certain user
- start any once played song again via id
- search in title from played songs
- (planned) combined search expressions
(un)subscribe to the Bob to hear music in any channel
(un)subscribe the Bob to certain channels
Playlist management for all users
broken | Basic plugin support

In progress:

Getting some really annoying bugs fixed :/

In planning:

Create multiple client instances automatically for diffrent channels
(Improved) Rights system
Own web-interface
Inteface API

Existing commands

All in all, the bot is fully operable only via chat (and actually only via chat).
Commands are invoked with !command.
Some commands have restrictions, like they can only be used in a private chat, only in public chat, or need admin rights.

add: Adds a new song to the queue.
clear: Removes all songs from the current playlist.
eval: Executes a given command or string
help: Shows all commands or detailed help about a specific command.
history: Shows recently played songs.
if: - no description yet -
kickme: Guess what?
link: Gets a link to the origin of the current song.
loop: Gets or sets whether or not to loop the entire playlist.
next: Plays the next song in the playlist.
parse: Displays the AST of the requested command.
pause: Well, pauses the song. Undo with !play
play: Automatically tries to decide whether the link is a special resource (like youtube) or a direct resource (like ./hello.mp3) and starts it
pm: Requests a private session with the ServerBot so you can invoke private commands.
previous: Plays the previous song in the playlist.
print: Lets you format multiple parameter to one.
quit: Closes the TS3AudioBot application.
quiz: Enable to hide the songnames and let your friends guess the title.
random: Gets whether or not to play playlists in random order.
repeat: Gets or sets whether or not to loop a single song.
rng: Gets a random number.
seek: Jumps to a timemark within the current song.
song: Tells you the name of the current song.
stop: Stops the current song.
subscribe: Lets you hear the music independent from the channel you are in.
take: Take a substring from a string
test: Only for debugging purposes
unsubscribe: Only lets you hear the music in active channels again.
volume: Sets the volume level of the music.

If the bot can't play some youtube videos it might be due to some embedding restrictions, which are blocking this.
You can add a youtube-dl binary or source folder and specify the path in the config to try to bypass this.

How to set up the bot

Dependencies

Any C# Compiler (Visual Studio or mono 4.0.0+ and xbuild)
(Linux only) A C Compiler for Opus

Compilation

Download the git repository with git clone https://github.com/Splamy/TS3AudioBot.git.
If you know what you are doing you can alternatively compile each depenency referenced here from source/git by yourself, but I won't add a tutorial for that.

Linux

See if you have NuGet by just executing nuget. If not, get NuGet.exe with wget https://dist.nuget.org/win-x86-commandline/latest/nuget.exe
Go into the directory of the repository with cd TS3AudioBot
Execute nuget restore or mono ../Nuget.exe restore to download all dependencies
Execute xbuild /p:Configuration=Release TS3AudioBot.sln to build the C# AudioBot
Make the Opus script runnable with chmod u+x InstallOpus.sh and run it with ./InstallOpus.sh
Get the ffmpeg 32bit or 64bit binary.
Extract the ffmpeg archive with tar -vxf ffmpeg-git-XXbit-static.tar.xz
Get the ffmpeg binary from ffmpeg-git-*DATE*-64bit-static\ffmpeg and copy it to TS3AudioBot/bin/Release/

Windows

Build the C# AudioBot with Visual Studio.
Download the latest libopus file from here
Open the archive and copy the file from opusfile-0.7-win32\libopus-0.dll, move and rename it to TS3AudioBot\bin\Release\libopus.dll
Get the ffmpeg 32bit or 64bit binary.
Open the archive and copy the ffmpeg binary from ffmpeg-latest-winXX-static\bin\ffmpeg.exe to TS3AudioBot\bin\Release\

Installation

Create 2 groups on the TeamSpeak server:

one for the ServerBot with enough rights so he can
- join as a serverquery
- view all server/channel/clients
- write in all/private chat
- optionally kick clients form channel/server
- See here for a more detailed list until I update the readme...
one for the AudioBotAdmin with no requirements (just ensure a high enough i_group_needed_member_add_power).

The first time you'll need to run the TS3Audiobot.exe without parameter and it will ask you a few questions. You can get ServerGroupIds in the rights window.
Now you can move the process to the backgroud or close the bot with !quit in teamspeak and run it in the background.
The recommended start from now on is TS3AudioBot.exe -q to disable writing to stdout since the bot logs everything to a log file anyway.
Congratz, you're done! Enjoy listening to your favourite music, experimenting with the crazy command system or do whatever you whish to do ;).

Testing and Fuzzying

Run the TS3ABotUnitTests project in Visual Studio or Monodevelop.

robotrono/TS3AudioBot