Schema Extensions

Schema extensions add gateway-level type definitions and resolvers into a combined API, which is useful for establishing connections between types that exist in separate subgraphs.

When considering these capabilities, be sure to compare them with the newer automated features available through type merging. While type merging frequently eliminates the need for schema extensions, it does not preclude their use.

Motivational Example

Let’s say we have a subgraph A, and the supergraph like below;

type Repository {
  id: ID!
  url: String
  issues: [Issue]
  userId: ID!
}
 
type Issue {
  id: ID!
  text: String!
  repository: Repository!
}
 
type Query {
  repositoryById(id: ID!): Repository
  repositoriesByUserId(id: ID!): [Repository]
}

And together with other composed subgraphs or extensions, we have the following supergraph;

type Repository {
  id: ID!
  url: String
  issues: [Issue]
  userId: ID!
  user: User
}
 
type Issue {
  id: ID!
  text: String!
  repository: Repository!
}
 
type User {
  id: ID!
  username: String
  repositories: [Repository]
    @resolveTo(
      sourceName: "A"
      sourceTypeName: "Query"
      sourceFieldName: "repositoriesByUserId"
      requiredSelectionSet: "{ id }"
      sourceArgs: { id: "{root.id}" }
    )
}
 
type Query {
  userById(id: ID!): User
}

Suppose we want the supergraph to delegate retrieval of repositories to the subgraph A, in order to execute queries such as this one:

query {
  userById(id: "1") {
    id
    username
    repositories {
      id
      url
      user {
        username
        id
      }
      issues {
        text
      }
    }
  }
}

The resolver function for the repositories field of the User type would be responsible for the delegation, in this case. While it’s possible to call a remote GraphQL endpoint or resolve the data manually, this would require us to transform the query manually or always fetch all possible fields, which could lead to overfetching. Delegation automatically extracts the appropriate query to send to the subgraph A:

# To Subgraph A
query ($id: ID!) {
  repositoriesByUserId(id: $id) {
    id
    url
    issues {
      text
    }
  }
}

The delegation also removes the fields that don’t exist on the subschema, such as user. This field would be retrieved from the parent schema using normal GraphQL resolvers.

Each field on the Repository and Issue types use a special resolver to properly extract data from the delegated response. The special resolver resolves aliases, converts custom scalars and enums to their internal representations, and maps errors. So it is more than just a simple fetch call, and return that.

Basic Example

Let’s say we have two subgraphs;

type Post {
  id: ID!
  text: String
  userId: ID!
}
 
type Query {
  postById(id: ID!): Post
  postsByUserId(userId: ID!): [Post!]!
}

type User {
  id: ID!
  email: String
}
 
type Query {
  userById(id: ID!): User
}

We may want to navigate from a particular user to their posts, or from a post to its user. This is possible within our service architecture by connecting an existing key of each object to a corresponding root query:

• Post.userId -> userById(id) gets a Post’s user.
• User.id -> postsByUserId(userId) gets a User’s posts.

To formalize this navigation within our supergraph, we can extend each type with a new field that will translate its respective key into an actual object association:

import { defineConfig } from '@graphql-mesh/compose-cli'
import { loadGraphQLHTTPSubgraph } from '@graphql-mesh/graphql'
 
export const composeConfig = defineConfig({
  subgraphs: [
    {
      sourceHandler: loadGraphQLHTTPSubgraph('Posts', {
        endpoint: 'http://localhost:4001/posts'
      })
    },
    {
      sourceHandler: loadGraphQLHTTPSubgraph('Users', {
        endpoint: 'http://localhost:4002/users'
      })
    }
  ],
  additionalTypeDefs: /* GraphQL */ `
    extend type Post {
      user: User!
    }
    extend type User {
      posts: [Post!]!
    }
  `
})

The additionalTypeDefs option provides type extensions (using the extend keyword) that add additional fields into the combined gateway schema and therefore may cross-reference types from any subgraph.

However, these extensions alone won’t do anything until they have corresponding resolvers. A complete example would look like this:

import { defineConfig } from '@graphql-mesh/compose-cli'
import { loadGraphQLHTTPSubgraph } from '@graphql-mesh/graphql'
 
export const composeConfig = defineConfig({
  subgraphs: [
    {
      sourceHandler: loadGraphQLHTTPSubgraph('Posts', {
        endpoint: 'http://localhost:4001/posts'
      })
    },
    {
      sourceHandler: loadGraphQLHTTPSubgraph('Users', {
        endpoint: 'http://localhost:4002/users'
      })
    }
  ],
  additionalTypeDefs: /* GraphQL */ `
    extend type Post {
      user: User!
        @resolveTo(
          sourceName: "Users"
          sourceTypeName: "Query"
          sourceFieldName: "userById"
          requiredSelectionSet: "{ id }"
          sourceArgs: { id: "{root.userId}" }
        )
    }
    extend type User {
      posts: [Post!]!
        @resolveTo(
          sourceName: "Posts"
          sourceTypeName: "Query"
          sourceFieldName: "postsByUserId"
          requiredSelectionSet: "{ id }"
          sourceArgs: { userId: "{root.id}" }
        )
    }
  `
})

When resolving User.posts and Post.user, we delegate each key reference to its corresponding root query.

• sourceName specifies the subgraph to delegate to.
• sourceTypeName and sourceFieldName specify the root query to call.
• sourceArgs specifies the arguments to be passed to the delegated field
• requiredSelectionSet above specifies the key field(s) needed from an object to query for its associations. For example, Post.user will require that a Post provide its userId. Rather than relying on incoming queries to manually request this key for the association, the selection set will automatically be included in subgraph requests to guarantee that these fields are fetched.

• returnType specifies the expected return type of the delegated operation. This is necessary when the return type of the delegated operation differs from the type of the field being resolved. For example, if Post.user resolves to a User type, but the delegated operation returns a UserResponse type, then returnType should be set to User.
• result specifies the path to the data in the response that should be returned. This is necessary when the data is nested in the response. For example, if the response is { data: { user: { id: 1, name: "John" } } }, then result should be set to data.user.

Batch Delegation (Array Batching)

The drawback of performing individual resolveTo calls is that they can be fairly inefficient. Say we request Post.user from an array of ten posts—that would delegate ten individual userById queries while resolving each user! To improve this, we can instead delegate in batches, where many instances of a field resolver are consolidated into one delegation.

To setup batching, the first thing we’ll need is a new query in the users’ service that allows fetching many users at once:

usersByIds(ids: [ID!]!): [User]!

With this many-users query available, we can now delegate the Post.user field in batches across many records:

import { defineConfig } from '@graphql-mesh/compose-cli'
import { loadGraphQLHTTPSubgraph } from '@graphql-mesh/graphql'
 
export const composeConfig = defineConfig({
  subgraphs: [
    {
      sourceHandler: loadGraphQLHTTPSubgraph('Posts', {
        endpoint: 'http://localhost:4001/posts'
      })
    },
    {
      sourceHandler: loadGraphQLHTTPSubgraph('Users', {
        endpoint: 'http://localhost:4002/users'
      })
    }
  ],
  additionalTypeDefs: /* GraphQL */ `
    extend type Post {
      user: User!
        @resolveTo(
          sourceName: "Users"
          sourceTypeName: "Query"
          sourceFieldName: "usersByIds"
          keyField: "userId"
          keysArg: "ids"
        )
    }
  `
})

If the query you’re delegating to don’t conform to these expectations, you can provide a custom resolver function in additionalResolvers to the Hive Gateway runtime configuration.

Batch delegation is generally preferable over plain delegation because it eliminates the redundancy of requesting the same field across an array of parent objects. Even so, delegation costs can add up because there is still one subschema request made per batched field—for remote services, this may create many network requests sent to the same service.

Hiding internal fields

Let’s say if you don’t want to expose usersByIds, you can use Filter Transform to hide it from the final schema. This is helpful when you want to stitch subgraphs but you don’t want to expose some extra fields that should not be exposed to the client.

import { createFilterTransform, defineConfig } from '@graphql-mesh/compose-cli'
import { loadGraphQLHTTPSubgraph } from '@graphql-mesh/graphql'
 
export const composeConfig = defineConfig({
  subgraphs: [
    {
      sourceHandler: loadGraphQLHTTPSubgraph('Posts', {
        endpoint: 'http://localhost:4001/posts'
      })
    },
    {
      sourceHandler: loadGraphQLHTTPSubgraph('Users', {
        endpoint: 'http://localhost:4002/users'
      }),
      transforms: [
        createFilterTransform({
          fieldFilter: (typeName, fieldName) => fieldName !== 'usersByIds'
        })
      ]
    }
  ],
  additionalTypeDefs: /* GraphQL */ `
    extend type Post {
      user: User!
        @resolveTo(
          sourceName: "Users"
          sourceTypeName: "Query"
          sourceFieldName: "usersByIds"
          keyField: "userId"
          keysArg: "ids"
        )
    }
  `
})

Schema Registry / Hive

You can use Schema Extensions with Hive. Since we don’t have a composition configuration of supergraph when we use a Hive project, we have to use Hive’s settings to configure additionalTypeDefs.

All you need to do is to provide the additionalTypeDefs in the Base Schema section of your project’s Settings tab, like below.