Replies: 1 comment
-
The whole principle of tRPC is that it is schemaless with the schema being transient. It sort of requires you to have the dependent services in the same project so it's not really designed for public-facing APIs. That said, there are discussions about it over at #755 and #271 |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Say I have a monorepo with two clients using the same server. There is a dependency from the clients to the entire server implementation.
https://github.com/KATT/zart/blob/60e7b37445f442e3eafec22f24bb069ca076e461/apps/nextjs/src/pages/api/trpc/%5Btrpc%5D.ts#L5-L6
Is there a way to avoid a hard dependency on the router impl? In its defense though, because clients and servers live in same codebase this becomes less of an issue than it was traditionally.
GraphQL achieves this by using a schema, although I will admit that
graphql-code-generator
becomes clunky to use, and GQL is designed around language interoperability, whereas trpc focuses on everything as TS for now.Just interested in thoughts around this.
Beta Was this translation helpful? Give feedback.
All reactions