Dataglot

Rust-native federated SQL query engine with governance enforced in the query plan. One PostgreSQL endpoint for all your databases.

Dataglot speaks the PostgreSQL wire protocol, so psql, DBeaver, Metabase, Tableau, dbt, and every Postgres driver connect with no special client. Point it at the databases you already run — PostgreSQL, MySQL, Oracle, Snowflake, S3/Parquet/CSV, Apache Iceberg — and query across all of them in one SQL statement, with column masks and row filters compiled into the plan, not bolted on after it.

Website & docs: dataglot.ai

Status: pre-1.0 and evolving. Dataglot is a read-path federation and governance engine, not a general-purpose PostgreSQL replacement. See known limitations.

Install

Prebuilt everywhere — no toolchain, no compiling (full menu, including tarballs and cargo binstall, in docs/install.md). Pick one channel:

Homebrew (macOS / Linux):

brew install dataglotai/tap/dataglot

Container image (published on loopback only — the default auth mode trusts any local user):

docker run --rm -p 127.0.0.1:5432:5432 ghcr.io/dataglotai/dataglot:latest

Quick start

QUICKSTART.md takes you from install to a governed, federated query in about five minutes — sources, credentials, and policies are all administered over psql with plain SQL (docs/runtime-config.md).

You don't need a server to try it — the binary runs a query in-process using the same federation and plan-time governance the server applies:

dataglot query "SELECT 1 AS n, 'hello' AS greeting"
dataglot query -c dataglot.toml "SELECT * FROM pg.public.users LIMIT 5"
dataglot shell -c dataglot.toml      # interactive REPL; \q to quit

To build from source instead:

# Prerequisites: Rust 1.94+ (workspace MSRV), nightly toolchain, cargo-deny + taplo
rustup toolchain install nightly
cargo install cargo-deny taplo-cli

# Build and test
make ci

Crates

Crate	Purpose
dataglot-core	Shared types, SessionContextFactory, error types
dataglot-federation	Postgres, MySQL, warehouse, and lakehouse connectors via datafusion-federation and iceberg-datafusion
dataglot-pgwire	PostgreSQL wire protocol via pgwire + datafusion-postgres
dataglot-policy	Policy enforcement — column masking, row filtering, tag-based governance
dataglot-catalog	Catalog metadata and the embedded meta-store
dataglot-server	Binary entrypoint, config, CLI, dashboard
dataglot-ballista	Optional distributed execution on Apache Ballista
dataglot-test-support	Shared test helpers for the workspace's unit tests

Docs map

Doc	What it gives you
QUICKSTART.md	Install → first governed federated query in ~5 minutes
docs/getting-started.md	The same journey, in more depth
docs/install.md	Every install channel: Homebrew, Docker, tarballs, cargo binstall, from source
docs/quick-reference.md	One-page cheat-sheet: build commands, features, runtime flags, dashboard URLs, auth modes
docs/configuration.md	Full dataglot.toml reference: catalogs, policies, TLS, auth, rate limits
docs/runtime-config.md	SQL-native administration: CREATE CATALOG, CREATE MASK, secrets
docs/access-control.md	Governance model: masks, row filters, tags, identities
docs/authentication.md	Auth modes and hardening before exposing a port
docs/compatibility.md	What works, what doesn't, and how far Postgres compatibility goes
CONTRIBUTING.md	PR workflow, branch/commit conventions, crate boundaries
SECURITY.md	Vulnerability reporting policy

License

Apache-2.0