trpc-router

To avoid sequential database queries in tRPC routers, use Promise.all within a single detail procedure to fetch related data in parallel. This reduces client round-trips and sequential request overhead, creating an aggregated detail endpoint.

Inject database models into tRPC procedures using middleware. This ensures models are created once per request and attached to the context (ctx), providing consistent model injection across all queries and mutations without repetitive boilerplate.

The best way to handle errors and validate inputs in a tRPC server API is by wrapping errors with TRPCError and validating inputs using Zod schemas. This enforces standardized return shapes and reliable server-side endpoint logic.

Create a tRPC procedure with Zod validation by defining schema inputs and enforcing standardized return shapes. Apply this clean procedure pattern to queries and mutations to ensure predictable, safe, and efficient server-side endpoint logic.

Yes, this tRPC router guide works directly with existing lambda router files. It is specifically applicable for creating or modifying routers and endpoint logic located in the src/server/routers/lambda directory using middleware-injected ctx models.