Base Repository Methods (API)

The Database Layer of the Boilerplate enforces the Repository Pattern, serving as the central hub for database interactions via Drizzle ORM.

By passing all Actions through this layer, we guarantee standardized connection pooling, global Query Logging, and simplified query syntaxes.

Drizzle Manager

The heart of the module exported by src/infrastructure/repositories/repository.ts is the pre-configured manager instance.

Instead of dealing with client connection life-cycles, you just import the repository and write semantic SQL immediately. It utilizes a highly concurrent postgres-js pool under the hood.

import repository from "@infrastructure/repositories/repository";
import entity from "../entity";
import { eq } from "drizzle-orm";

// Finding a single record
const result = await repository
  .select()
  .from(entity)
  .where(eq(entity.id, "123"))
  .limit(1);

// Inserting data
await repository
  .insert(entity)
  .values({ email: "hello@world.com" })
  .returning();

// Transaction blocks
await repository.transaction(async (tx) => {
  await tx.insert(entity).values({/* ... */});
  await tx.update(anotherEntity).set({/* ... */});
});

TIP

For a deep dive into Cache Interception techniques when querying through the Repository, refer to the Repository Pattern Concept documentation.

Utilities

withPagination

When querying large lists of resources, building LIMIT and OFFSET manually is repetitive. This built-in helper applies offset-based pagination directly to the active QueryBuilder.

Signature: withPagination(qb: PgSelect, page: number, size?: number)

By default, the active page size (size) is set to 10.

import { withPagination } from "@infrastructure/repositories/repository";
import repository from "@infrastructure/repositories/repository";
import entity from "../entity";

const page = 3; 

const pagedQuery = withPagination(
  repository.select().from(entity),
  page,
  20 // 20 items per page
);

const results = await pagedQuery.execute();