Ship162 Plugin

The tangram_ship162 plugin integrates AIS data from a ship162 instance, enabling real-time visualization and historical analysis of maritime traffic.

Overview

• a background service to maintain a real-time state of all visible ships and persist their data.
• a rest api endpoint (/data/ship/{mmsi}) to fetch the full, time-ordered trajectory for a specific ship.
• frontend components to render ships on the map and display detailed information.

sequenceDiagram
    participant S as ship162
    participant R as Redis
    participant P as ships service
    participant C as channel service
    participant F as Frontend

    S->>R: PUBLISH ship162 (raw message)
    P->>R: SUBSCRIBE ship162
    Note over P: Process message, update state vector
    P->>R: XADD history:ingest:ship162, *, ...

    loop every second
        Note over P: Filter ships by client's bbox
        P->>R: PUBLISH to:streaming-client1:new-ship162-data
    end

    C->>R: SUBSCRIBE to:streaming-client1:*
    C->>F: PUSH new-ship162-data

Redis Events

Direction	Channel	Event/Command	Payload
Input	ship162	PUBLISH	Raw JSON message from ship162.
Output	to:streaming-{id}:new-ship162-data	PUBLISH	{ "count": 123, "ship": [...] } containing visible ships.
Output	history:ingest:ship162	XADD	Apache Arrow record batch (binary).

Configuration

To use this plugin, you must have a running ship162 instance publishing data to redis.

tangram.toml

[core]
plugins = ["tangram_ship162", "tangram_history"]

[plugins.tangram_ship162]
# redis channel that ship162 is publishing to.
ship162_channel = "ship162"

# how long (in seconds) to keep a ship in the state vector table
# after its last message.
state_vector_expire = 600

# history persistence settings
history_table_name = "ship162"
history_flush_interval_secs = 5
history_buffer_size = 100000