Skip to content

Latest commit

 

History

History
91 lines (67 loc) · 8.48 KB

README.md

File metadata and controls

91 lines (67 loc) · 8.48 KB

lilith_tf2e

lilith_tf2e is intended to be a fully-fledged helper tool for interacting with, and recording data from, Team Fortress 2.

In line with the TF2 Bot Kicker GUI by Bash09, and the plans behind MegaScatterBomb's MegaAntiCheat and associated helper clients, this is intended to be used by legitimate players to aid with investigation, recording and reporting suspicious players and bots. But this app also aims to be a lot more - it aims to be an effective data collection tool for recording player interactions, stats, scores and games, as well as automatically recording demos.

It will implement a robust data archival system involving compression algorithms (probably python gzip library) and an automatic repository upkeep tool to minimise sparse, out of date, and/or unwanted data being collected.

The design

lilith_tf2e intends to be 100% Python, end-to-end. It uses PySimpleGUI for the user interface (because I wanted to learn it, but the other alternative would be to turn this into a Flask server and display it via a web-browser or electron app). It will use SQLite3 for databasing and data archival. It will use pythonic job schedulers and the python package 'schedule' to run ephemeral/sparse upkeep and maintenance jobs.

The timeline

This table will be added to as tasks are planned and as each task evolves.

Task Name Start Time End Time Stage Current Focus Issues Result
Bespoke Path Listener/Watcher prior to go-live prior to go-live Complete. - Existing Pythonic packages do not work to solve the task at hand. clwd package
Interface with RCON and Steam API prior to go-live prior to go-live Complete (with issues). Refactor code and packaging - rc package
Simple GUI App for PoC prior to go-live 25/05/2023 Complete. - - gui/lobbyviewer.py app
Refactor and reform code base for launch platform for MVP 25/05/2023 - In progress. - - -
Implement G15 parser to parse g15_dumpplayer command output 30/06/2023 30/06/2023 Complete. integrating the parser with existing UI and lobby code to leverage it g15_dumpplayer returns data via RCON, yay! Downside - the packets are fragmented because the output is massive. Need to implement proper packet de-frag g15parser package, FragClient rcon client
Interface console listener (l2) for lobby updates 25/05/2023 - In progress. Develop effective startup sequence that avoids desyncs and unnecessary RCON commands
Convert this to a poetry project for dependency management - - Planned. - - -
Minimum viable product for full product - - Planned. - - -

Current in progress

Cleaning up and preparing for 3rd refactor of the UI code for the g15 viewer (simple lobby viewer class that interprets g15_dumpplayer data). Looking into grabbing PySimpleGui_Events to handle app events better. Need to start implementing classes in the GUI packages to encapsulate data better.

(NEW!) G15 Comic Lobby Viewer Example: Functioning comical g15 lobby viewer in a random 2fort lobby

(TEST) G15 Lobby Viewer Example: Functioning G15 lobby viewer from a random tf2 lobby i was in

(OLD (OLD)) Simple Lobby Viewer Example: Functioning simple lobby viewer from a random TF2 lobby I was in

NOTE:

You will need to make a data/ directory to house a config.yml and a .env file. They can be structured as such:

config.yml:

tf2:
  dir: "path/to/tf2"
  rcon:
    password: "lilith_is_hot"

and

.env:

STEAM_WEB_API_KEY="YOUR_STEAM_API_KEY_HERE"

I also used two photos in the data/ dir that get loaded by the gui/lobbyviewer.py. These were lilith.png and tf2.png. I have NOT included these in the repo because I do not own the TF2 logo (although it would probably be fine if I used it), and because Lilith is my OC and the artwork is commissioned and not MIT licensed.

These images were 2000x2000 and in PNG format. Feel free to replace them with whatever you like.

Main dependencies

  • PySimpleGUI
  • steam_converter
  • rcon
  • steam
  • psutil
  • schedule
  • loguru
  • PyYAML

My intention is for this application to be fairly heavy weight, but nicely concurrent with a mix of python multiprocessing and threading.

I had to add the following lines to my autoexec.cfg to allow the gui/lobbyviewer.py to interface with TF2:

ip 0.0.0.0
rcon_password lilith_is_hot
net_start

Note that you don't actually need to specify the rcon_password as such, but rather it can be whatever you want, as long as data/config.yml corresponds to this.

The purpose

This is intended to be a 100% Python full-stack application to interface with Team Fortress 2. This will be achieved via RCON and monitoring the console.log file, and will take advantage of the Steam Web API, and several other external APIs (such as Bash09's cheater list, and megascatterbomb's mcd or anti-cheat (when its released)).

If you want an actually complete and functional program, see Bash's TF2 Bot Kicker GUI (written in Rust! blazingly fast!) --> https://github.com/Bash-09/tf2-bot-kicker-gui