rrrocket
rrrocket parses a Rocket League replay file and outputs JSON.
Underneath rrrocket is the general rocket league parsing library: boxcars
Installation
Download the appropriate bundle from the releases page:
- For Windows, you'll want the one labeled
windows-msvc - For Linux, you'll want the one labeled
linux-musl - For macOS, you'll want the only one labeled
apple
Usage
Parses Rocket League replay files and outputs JSON with decoded information
USAGE:
rrrocket [FLAGS] [input]...
FLAGS:
-n, --network-parse parses the network data of a replay instead of skipping it
-c, --crc-check forces a crc check for corruption even when replay was successfully parsed
--dry-run parses but does not write JSON output
-h, --help Prints help information
-j, --json-lines output multiple files to stdout via json lines
-m, --multiple parse multiple replays in provided directories. Defaults to writing to a sibling JSON file,
but can output to stdout with --json-lines
-p, --pretty output replay as pretty-printed JSON
-V, --version Prints version information
ARGS:
<input>... Rocket League replay files
Output
A sample output of the JSON from rrrocket:
{
"header_size": 4768,
"header_crc": 337843175,
"major_version": 868,
"minor_version": 12,
"game_type": "TAGame.Replay_Soccar_TA",
"properties": {
"TeamSize": 3,
"Team0Score": 5,
"Team1Score": 2,
"Goals": [
{
"PlayerName": "Cakeboss",
"PlayerTeam": 1,
"frame": 441
},
// all the goals
]
// and many more properties
}If network parsed is enabled then an attribute (snipped) looks something like:
{
"actor_id": 6,
"stream_id": 51,
"attribute": {
"RigidBody": {
"sleeping": true,
"location": {
"bias": 16384,
"dx": 16384,
"dy": 16384,
"dz": 25658
},
"x": 1,
"y": 1,
"z": 1,
"linear_velocity": null,
"angular_velocity": null
}
}
}Queries with jq
Since rrrocket outputs json, jq is a natural query tool. Here are some questions that rrrocket and jq can answer together.
Want to find out which non 3v3 games had a score difference greater than 2?
rrrocket --json-lines --multiple ~/projects/boxcars/assets/replays/good/ \
| jq -c 'if (.replay.properties.TeamSize != 3 and
(((.replay.properties.Team0Score // 0) - (.replay.properties.Team1Score // 0) | length) > 2)) then .file else empty end'
Top combined score?
rrrocket --json-lines --multiple ~/projects/boxcars/assets/replays/good/ \
| jq -c '{(.file): (.replay.properties.Team0Score // 0 + .replay.properties.Team1Score // 0)}' \
| sort -n -k2 -t ':'
Games with certain attributes?
rrrocket --json-lines --multiple ~/projects/boxcars/assets/replays/good/ \
| jq -c 'if (.replay.objects | contains(["Archetypes.Ball.Ball_Breakout"])) then .file else empty end'
boxcapy
boxcapy is a python script that ingests given JSON files that have been created
by rrrocket (and only needs header information). The command below is the one
I use to generate the JSON files:
find . -type f -iname "*.replay" | xargs ~/rrrocket -m To have your graphs saved into your directory follow the below instructions:
- Since the graphs are in the style of XKCD, one has to install the Humor Sans font before continuing (eg.
apt install fonts-humor-sans) - Install pipenv
- Install dependencies
pipenv && pipenv install - Run on generated JSON files:
pipenv run boxcapy/rocket-plot.py *.json --headless