By putting the loader inside the client.js, we're sending all the schema creation logic to the client, which is not necessary. Furthermore, the only reason stuff like introspectionQuery is needed is if someone is getting the schema remotely, which I assume is a minority of folks.
We gotta come up with a better way to keep compile-time logic outside of the client payload.
Starting point: run webpack stats & see what depends on transformSchema
By putting the loader inside the
client.js, we're sending all the schema creation logic to the client, which is not necessary. Furthermore, the only reason stuff likeintrospectionQueryis needed is if someone is getting the schema remotely, which I assume is a minority of folks.We gotta come up with a better way to keep compile-time logic outside of the client payload. Starting point: run webpack stats & see what depends on
transformSchema