We now have a discord channel: Discord

Demoparser is a tool for analysing CS2 replay files "demos". All the heavy lifting is done in Rust, but you use it from the comfort of Python/JavaScript. The parser takes a slightly different approach to exposing the information in the demo. Rather than letting you hook onto events in a streaming fashion, it lets you "query" the demo similarly to how you would interact with a database.

Python: pip install demoparser2 (python >= 3.8)

NodeJS: npm i @laihoe/demoparser2

WASM: npm i demoparser2

from demoparser2 import DemoParser

parser = DemoParser("path_to_demo.dem")
event_df = parser.parse_event("player_death", player=["X", "Y"], other=["total_rounds_played"])
ticks_df = parser.parse_ticks(["X", "Y"])

var {parseEvent, parseTicks} = require('@laihoe/demoparser2');

let event_json = parseEvent("path_to_demo.dem", "player_death", ["X", "Y"], ["total_rounds_played"])
let ticks_json = parseTicks("path_to_demo.dem", ["X", "Y"])

• Examples

• Documentaion

The benchmark finds the coordinates of all player deaths. Code: Benchmark

Benchmarked 50 demos containing a mix of (MM, Faceit, HLTV) demos. Total size: 4.6GB

Gaming pc specs: CPU: Ryzen 5900x, SSD: Samsung 980 pro NVME
Thinkpad specs: CPU: i5-1335g7, SSD: Toshiba XG6 NVME
Both devices run Ubuntu 20.04
Python/JS are rougly as fast.

True/Flase if player is pressing button.

(buttons is the real value of m_nButtonDownMaskPrev and the others are derived from it)

Go: https://github.com/markus-wa/demoinfocs-golang
C#: https://github.com/saul/demofile-net
Python: https://github.com/pnxenopoulos/awpy
Java: https://github.com/skadistats/clarity

Without Dotabuff's dota 2 parser "manta" this would not have been possible. Check it out: https://github.com/dotabuff/manta

The dota 2 demo format is very similar to CS2 demo format with only a few minor changes.

Thank you to Clarity for an elegant implementation of sendtables

Thank you to Dandrews for figuring out voice data extraction