ArkView

ArkView is a client-side plugin for the Arkon bot that lets you view your map data in real time. It runs on your server, parses the save, and serves the data over HTTP for Arkon to consume.

ArkView processes one map per instance — run multiple maps via multiple instances (one container / one systemd unit / one exe per map).

How it works

Parsing is handled by the pure-Python arkparser library — no .NET, no external exe. It reads both engine generations:

• ASE (Survival Evolved) — .ark flat-binary saves
• ASA (Survival Ascended) — *_WP.ark SQLite saves

Each parse runs in a short-lived child process that writes ASV_*.json and exits, so the multi-GB parse footprint is reclaimed by the OS and the long-lived server stays lean. There is no database: results are written to an output directory and held in memory until the next parse. ArkView watches the map file's modification time and re-parses (every ~5s) when the server saves.

Migrating from the old (.NET) build? The HTTP API and response shape are unchanged, so the Arkon/AVClient side needs no changes. You no longer need the .NET runtime or the bundled ASVExport.exe.

Configuration

ArkView reads a config.ini (created on first run; or copy default_config.ini).

[Settings]
# Port for the API to listen on (TCP). Forward it in your router/firewall.
Port = 8000

# Direct path to the .ark map file.
# ASE: .../Saved/SavedArks/TheIsland.ark
# ASA: .../Saved/SavedArks/TheIsland_WP/TheIsland_WP.ark
# Player (.arkprofile) and tribe (.arktribe) files are read from the same directory.
MapFilePath = path/to/your/map.ark

# (Optional) Direct path to the cluster/solecluster folder
ClusterFolderPath = path/to/your/cluster

# (Optional) Direct path to BanList.txt
BanListFile = path/to/your/BanList.txt

# Parser child priority (Windows): LOW, BELOWNORMAL, NORMAL, ABOVENORMAL, HIGH
Priority = LOW

# CPU cores the parser child may use (capped to the host core count)
Threads = 2

# Re-parse when cluster files change (not just the map file)
ReprocessOnArkDataUpdate = False

# If true, the API binds to 127.0.0.1 only
Debug = False

# (Optional) Sentry DSN for error tracking
DSN =

# (Optional) API key for Bearer-token auth
APIKey =

Two env vars let one install serve many maps (see systemd below):

• ARKVIEWER_CONFIG — path to this instance's config.ini
• ARKVIEWER_OUTPUT — directory this instance writes ASV_*.json to

Running on Windows (.exe)

1. Download the latest ArkViewer.exe from Releases.
2. Run it anywhere — it creates a config.ini next to the exe on first launch.
3. Set MapFilePath (and optionally ClusterFolderPath), pick a Port, and forward that port. No .NET required.

Running from source (Python 3.10+)

git clone https://github.com/vertyco/arkview.git
cd arkview
python -m venv .venv
source .venv/bin/activate          # Windows: .venv\Scripts\Activate.ps1
pip install -r requirements.txt
cp default_config.ini config.ini   # then edit MapFilePath etc.
python main.py

Running with Docker

One container per map; no database or state volume (output is ephemeral and re-parsed on restart).

cp docker-compose.example.yml docker-compose.yml   # edit binds/ports/configs
docker build -t arkviewer:latest .
docker compose up -d

Each service binds a per-map config.ini at /app/config.ini and mounts the save tree read-only. See docker-compose.example.yml for the full layout. If your save-mount owner isn't UID 1000, build with --build-arg UID=<uid> GID=<gid>.

Running on Ubuntu (systemd)

A templated unit at deploy/arkview@.service runs one shared install as many instances. Assuming code + venv at /opt/arkviewer and per-map files under /opt/arkviewer/maps/<name>/:

sudo cp deploy/arkview@.service /etc/systemd/system/
sudo systemctl daemon-reload
sudo systemctl enable --now arkview@theisland.service
journalctl -u arkview@theisland -f

The unit sets ARKVIEWER_CONFIG/ARKVIEWER_OUTPUT per instance and caps memory at 4 GB (the parser child peaks ~2 GB on a large map — raise MemoryMax for very large/modded maps).

Adding to Arkon

Once ArkView is running:

• Type +viewservers to open the server menu (cluster + servers already added).
• Select your cluster and click Servers.
• Click ArkView and enter the port you're running on, then Submit (IP is usually auto-detected from your server's IP).

Credits

Parsing is powered by arkparser. The exported data schema descends from the ASV lineage (miragedmuk's ASV) — lots of love for the groundwork there.

Contributing

Found a bug or have a suggestion? Open an issue. For larger contributions, please reach out on Discord before opening a PR.