Skip to content

Getting Started

Joel Guerra edited this page Jun 11, 2026 · 2 revisions

Build SnoutDB

Install Odin, clone the repository, and build the CLI:

git clone https://github.com/jacovinus/snoutdb.git
cd snoutdb
odin build ./cmd/snout -out:snout
./snout version

Inspect a File

./snout csv-info data.csv
./snout jsonl-info events.jsonl
./snout log-info access.log

Explore Unknown Data

sniff infers column roles, reports statistics, and suggests queries:

./snout sniff -f data.csv
./snout sniff -f events.jsonl --format json
./snout sniff -f access.log

Run a Query

./snout -f data.csv group=region -- \
  avg=latency_ms p95=latency_ms count=rows \
  --sort p95=latency_ms desc \
  --limit 10

Create a .snout File

Import raw data once and reuse the typed snapshot:

./snout csv-import data.csv data.snout
./snout info data.snout
./snout -f data.snout group=region -- count=rows

Validate a Development Checkout

./scripts/validate-release.sh

This builds the CLI and C ABI, runs the complete Odin test suite, and exercises the primary ingestion, query, transformation, merge, and embedding workflows.

Clone this wiki locally