WrenAI/core/wren/README.md

wren-engine

Wren Engine CLI and Python SDK — semantic SQL layer for 20+ data sources.

Translate natural SQL queries through an MDL (Modeling Definition Language) semantic layer and execute them against your database. Powered by Apache DataFusion and Ibis.

Installation

pip install wren-engine              # Core (DuckDB included)
pip install wren-engine[postgres]    # PostgreSQL
pip install wren-engine[mysql]       # MySQL
pip install wren-engine[bigquery]    # BigQuery
pip install wren-engine[snowflake]   # Snowflake
pip install wren-engine[clickhouse]  # ClickHouse
pip install wren-engine[trino]       # Trino
pip install wren-engine[mssql]       # SQL Server
pip install wren-engine[databricks]  # Databricks
pip install wren-engine[redshift]    # Redshift
pip install wren-engine[spark]       # Spark
pip install wren-engine[athena]      # Athena
pip install wren-engine[oracle]      # Oracle
pip install 'wren-engine[memory]'    # Schema & query memory (LanceDB)
pip install 'wren-engine[all]'       # All connectors + memory

Requires Python 3.11+.

Quick start

1. Create ~/.wren/mdl.json — your semantic model:

{
  "catalog": "wren",
  "schema": "public",
  "models": [
    {
      "name": "orders",
      "tableReference": { "schema": "mydb", "table": "orders" },
      "columns": [
        { "name": "order_id",    "type": "integer" },
        { "name": "customer_id", "type": "integer" },
        { "name": "total",       "type": "double" },
        { "name": "status",      "type": "varchar" }
      ],
      "primaryKey": "order_id"
    }
  ]
}

2. Create ~/.wren/connection_info.json — your connection:

{
  "datasource": "mysql",
  "host": "localhost",
  "port": 3306,
  "database": "mydb",
  "user": "root",
  "password": "secret"
}

3. Run queries — wren auto-discovers both files from ~/.wren:

wren --sql 'SELECT order_id FROM "orders" LIMIT 10'

For the full CLI reference and per-datasource connection_info.json formats, see docs/cli.md and docs/connections.md.

4. Index schema for semantic search (optional, requires wren-engine[memory]):

wren memory index                              # index MDL schema
wren memory fetch -q "customer order price"    # fetch relevant schema context
wren memory store --nl "top customers" --sql "SELECT ..."  # store NL→SQL pair
wren memory recall -q "best customers"         # retrieve similar past queries

---

Python SDK

import base64, orjson
from wren import WrenEngine, DataSource

manifest = { ... }  # your MDL dict
manifest_str = base64.b64encode(orjson.dumps(manifest)).decode()

with WrenEngine(manifest_str, DataSource.mysql, {"host": "...", ...}) as engine:
    result = engine.query('SELECT * FROM "orders" LIMIT 10')
    print(result.to_pandas())

---

Development

just install-dev    # Install with dev dependencies
just lint           # Ruff format check + lint
just format         # Auto-fix

Command	What it runs	Docker needed
just test-unit	Unit tests	No
just test-duckdb	DuckDB connector tests	No
just test-postgres	PostgreSQL connector tests	Yes
just test-mysql	MySQL connector tests	Yes
just test	All tests	Yes

Publishing

./scripts/publish.sh            # Build + publish to PyPI
./scripts/publish.sh --test     # Build + publish to TestPyPI
./scripts/publish.sh --build    # Build only

License

Apache-2.0