📊 StatQL

StatQL is a desktop‑first SQL engine that streams fast, approximate answers from your data sources - without spinning up a server.

Initial release ships with connectors for PostgreSQL, Neo4j, Redis and the local Filesystem. More catalogs - cloud object storage, OLAP warehouses, REST APIs - are on the roadmap.

Quick start

To launch StatQL, install it and start up the local server:

pip install statql
python -m statql            # launches the Streamlit server

Then in your browser, navigate to http://localhost:8501

Integrate your data sources in the Integrations page, then start running your queries!

Usage examples

StatQL emits a fresh result table every 0.5 – 1 s. Each numeric cell is formatted as value ± error, where error is the absolute 95 % confidence‑interval half‑width. As more of the population is sampled, the estimates tighten.

1️⃣ Row count per order status (single table)

SELECT l_linestatus, COUNT() AS rows
FROM pg.local.tpch.public.lineitem
GROUP BY l_linestatus;

2️⃣ Total PNG footprint in /data/images (filesystem)

SELECT divide(SUM(size), 1_073_741_824) AS gib
FROM fs.media_fs.entries
WHERE get_file_ext(path) = 'png';

3️⃣ Row counts across all databases in a cluster (wildcards)

SELECT @db, COUNT() AS rows
FROM pg.us-cluster-4.?.public.lineitem
GROUP BY @db
ORDER BY rows DESC;

Video Tutorial

👉 Full docs → docs/index.md