Mutations

Mutation fields should use the @resolver directive to provide a field resolver that executes the mutation. For the following example schema:

extend type Mutation {
  publishListing(id: ID! @idOf(type: "Listing")): Listing @resolver
}

The resolver might look like:

@Resolver
class PublishListingResolver @Inject constructor(
  val client: ListingServiceClient
) : MutationResolvers.PublishListing() {
  override suspend fun resolve(ctx: Context): Listing {
    client.publish(ctx.arguments.id.internalID)
    return ctx.nodeFor(ctx.arguments.id) // Creates a Listing node reference
  }
}

As this example shows, resolvers for mutation fields are almost identical to query field resolvers. A major difference is that Context implements MutationFieldExecutionContext. This allows mutation field resolvers to execute submutations using Context.mutation() in addition to executing subqueries using Context.query().

Mutation field resolvers should still be annotated with @Resolver. However, they may not provide a required selection set using objectValueFragment, since those selections would include other mutation fields. Mutation field resolvers can execute other mutation fields by calling Context.mutation() instead.