prisma-archive/graphql-schema-cache

Shouldn't create a new schema per request

Closed this issue · 1 comments

As mentioned in #4 creating a new remote schema is expensive and shouldn't be done on a per request basis (or at least should be implemented more efficiently).

I'm caching the introspection schema result, the overhead of creating the actual schema from it is minimal, the introspection query takes most of the effort:
https://github.com/qewl/qewl/blob/master/src/middlewares/remoteSchema.ts#L68-L75