Configuration

Graph Lagoon Studio is configured via environment variables or programmatically through Settings.

Environment Variables

# Warehouse connection (local PySpark)
SQL_WAREHOUSE_URL=http://localhost:8001

# Database (optional — for persisting explorations/contexts)
GRAPH_LAGOON_DATABASE_ENABLED=true
GRAPH_LAGOON_DATABASE_URL=postgresql+asyncpg://user:pass@localhost:5432/sgraph

# Databricks mode (replaces local warehouse)
GRAPH_LAGOON_DATABRICKS_MODE=false
GRAPH_LAGOON_DATABRICKS_HOST=adb-xxx.azuredatabricks.net
GRAPH_LAGOON_DATABRICKS_TOKEN=dapi-xxx
GRAPH_LAGOON_DATABRICKS_WAREHOUSE_ID=xxx
GRAPH_LAGOON_DATABRICKS_CATALOG=main

# Development
GRAPH_LAGOON_DEV_MODE=true
GRAPH_LAGOON_SHOW_ERROR_DETAILS=true

Programmatic Configuration

from graphlagoon import Settings, create_mountable_app

settings = Settings(
    sql_warehouse_url="http://my-warehouse:8001",
    database_enabled=True,
    database_url="postgresql+asyncpg://user:pass@db:5432/graphs",
)

app = create_mountable_app(settings=settings)

Deployment Modes

Mode	Warehouse	Database	Command
Local dev (in-memory)	PySpark (local)	None	make dev
Local dev (persistent)	PySpark (local)	PostgreSQL	make dev-db
Databricks	Databricks SQL	None	make api-databricks
Databricks + local DB	Databricks SQL	PostgreSQL	make api-databricks-localdb
Embedded	Configured by host app	Configured by host app	pip install graphlagoon