Quick Start

This guide gets you from zero to a validated Spine data product using the Dataspine CLI. Spine is the .spine language for declaring types, ingests, outlets, and stream flows. Full reference: Spine language.

Prerequisites

• Dataspine CLI installed
• Basic familiarity with data pipelines (no Rust or SQL required for this walkthrough)

1. Create a Spine file

Create a directory and a single source file, for example spine-src/main.spine:

namespace com.example {
    Struct Event {
        message: String,
    }

    Ingest EventIngest {
        name: "EventIngest",
        type: Event,
    }

    Outlet EventOutlet {
        name: "EventOutlet",
        type: Event,
    }

    flow = EventIngest.map(e -> e).publish(EventOutlet)
}

This is a minimal pass-through map to one outlet, with a namespace suited to local experimentation.

2. Check and compile

From the project root (parent of spine-src/):

dataspine check --source-dir ./spine-src --root-namespace com.example

To emit a build artifact:

dataspine compile \
  --source-dir ./spine-src \
  --root-namespace com.example \
  --out ./build/artifact.json

Add --lib pointing at extra type libraries if your Spine references external types.

3. Next steps

• First Data Product Tutorial — ingest/expected JSON layout and a small extension to the pipeline.
• Data product lifecycle — design, test, and rollout with Spine.
• CLI reference — check, compile, auth and registry commands.