Skip to content

SQL

@kylobyte/keel/sql is Drizzle over @effect/sql-pg: a write-only repository builder, and a query layer that turns HTTP query params into filtered, sorted, paginated SQL. It needs three optional peers, and a fourth for the id helper:

Terminal window
pnpm add drizzle-orm@1.0.0-beta.22 @effect/sql@0.48.6 @effect/sql-pg@0.49.7
pnpm add snowyflake@2.0.1 # only if you use snowflakeId()

pg comes along as a dependency of @effect/sql-pg; install it directly only if your own code imports it — overriding type parsers, say.

Same deal as the runtime: keel does not own the connection. The app builds its own Drizzle Effect service — it decides which variable holds the URL, how the pool is configured, which pg type parsers are overridden — and hands the tag to createSql, which closes over it and returns the helpers. The tag then flows into the requirements of every repository built from it.

services/database/database.service.ts
import { PgClient } from "@effect/sql-pg";
import * as PgDrizzle from "drizzle-orm/effect-postgres";
import { Config, Effect } from "effect";
const PgClientLive = PgClient.layerConfig({
url: Config.redacted("DATABASE_URL"),
});
export class DatabaseService extends Effect.Service<DatabaseService>()(
"DatabaseService",
{
effect: PgDrizzle.make().pipe(Effect.provide(PgDrizzle.DefaultServices)),
dependencies: [PgClientLive],
},
) {}
shared/app/sql.ts
import { createSql } from "@kylobyte/keel/sql";
import { DatabaseService } from "../../services/database/database.service.ts";
export const { buildRepository } = createSql(DatabaseService);

The pages that follow are the reference for each piece. Adding a feature uses all of them at once, in order, to build one complete module — start there if you would rather see the shape before the details.