Federated schemas

Overview

A federated supergraph uses multiple "types" of GraphQL schemas:

Subgraph schemas. Each subgraph has a distinct schema that indicates which types and fields of your composed supergraph it can resolve.
- These are the only schemas that your teams define manually.
Supergraph schema. This schema combines all of the types and fields from your subgraph schemas, plus some federation-specific information that tells your router which subgraphs can resolve which fields.
- This schema is the result of performing composition on your collection of subgraph schemas.
API schema. This schema is similar to the supergraph schema, but it omits federation-specific types, fields, and directives that are considered "machinery" and are not part of your public API.
- This is the schema that your router exposes to clients, which don't need to know internal implementation details about your graph.

Let's look at an example!

Subgraph schemas

Below are example schemas for three subgraphs in an e-commerce company's supergraph. Each subgraph is implemented as a separate GraphQL API:

Users

type Query {
  me: User
}

type User @key(fields: "id") {
  id: ID!
  username: String! @shareable
}

# (Subgraph schemas include
# this to opt in to
# Federation 2 features.)
extend schema
  @link(url: "https://specs.apollo.dev/federation/v2.3",
        import: ["@key", "@shareable"])

Products

type Query {
  topProducts(first: Int = 5): [Product]
}

type Product @key(fields: "upc") {
  upc: String!
  name: String!
  price: Int
}

extend schema
  @link(url: "https://specs.apollo.dev/federation/v2.3",
        import: ["@key", "@shareable"])

Reviews

type Review {
  body: String
  author: User @provides(fields: "username")
  product: Product
}

type User @key(fields: "id") {
  id: ID!
  username: String! @external
  reviews: [Review]
}

type Product @key(fields: "upc") {
  upc: String!
  reviews: [Review]
}

# (This subgraph uses additional
# federated directives)
extend schema
  @link(url: "https://specs.apollo.dev/federation/v2.3",
        import: ["@key", "@shareable", "@provides", "@external"])

As these schemas show, multiple subgraphs can contribute unique fields to a single type. For example, the Products subgraph and the Reviews subgraph both contribute fields to the Product type.

Supergraph schema

The supergraph schema is the output of schema composition. It serves the following purposes:

It provides your router with the name and endpoint URL for each of your subgraphs.
It includes all types and fields defined by all of your subgraphs.
It tells your router which of your subgraphs can resolve which GraphQL fields.

Here's the supergraph schema composed with the subgraph schemas above:

1
schema
2
  @link(url: "https://specs.apollo.dev/link/v1.0")
3
  @link(url: "https://specs.apollo.dev/join/v0.3", for: EXECUTION)
4
{
5
  query: Query
6
}
7

8
directive @join__enumValue(graph: join__Graph!) repeatable on ENUM_VALUE
9

10
directive @join__field(graph: join__Graph, requires: join__FieldSet, provides: join__FieldSet, type: String, external: Boolean, override: String, usedOverridden: Boolean) repeatable on FIELD_DEFINITION | INPUT_FIELD_DEFINITION
11

12
directive @join__graph(name: String!, url: String!) on ENUM_VALUE
13

14
directive @join__implements(graph: join__Graph!, interface: String!) repeatable on OBJECT | INTERFACE
15

16
directive @join__type(graph: join__Graph!, key: join__FieldSet, extension: Boolean! = false, resolvable: Boolean! = true, isInterfaceObject: Boolean! = false) repeatable on OBJECT | INTERFACE | UNION | ENUM | INPUT_OBJECT | SCALAR
17

18
directive @join__unionMember(graph: join__Graph!, member: String!) repeatable on UNION
19

20
directive @link(url: String, as: String, for: link__Purpose, import: [link__Import]) repeatable on SCHEMA
21

22
scalar join__FieldSet
23

24
enum join__Graph {
25
  PRODUCTS @join__graph(name: "products", url: "http://localhost:4002/graphql")
26
  REVIEWS @join__graph(name: "reviews", url: "http://localhost:4003/graphql")
27
  USERS @join__graph(name: "users", url: "http://localhost:4001/graphql")
28
}
29

30
scalar link__Import
31

32
enum link__Purpose {
33
  """
34
  `SECURITY` features provide metadata necessary to securely resolve fields.
35
  """
36
  SECURITY
37

38
  """
39
  `EXECUTION` features provide metadata necessary for operation execution.
40
  """
41
  EXECUTION
42
}
43

44
type Product
45
  @join__type(graph: PRODUCTS, key: "upc")
46
  @join__type(graph: REVIEWS, key: "upc")
47
{
48
  upc: String!
49
  name: String! @join__field(graph: PRODUCTS)
50
  price: Int @join__field(graph: PRODUCTS)
51
  reviews: [Review] @join__field(graph: REVIEWS)
52
}
53

54
type Query
55
  @join__type(graph: PRODUCTS)
56
  @join__type(graph: REVIEWS)
57
  @join__type(graph: USERS)
58
{
59
  topProducts(first: Int = 5): [Product] @join__field(graph: PRODUCTS)
60
  me: User @join__field(graph: USERS)
61
}
62

63
type Review
64
  @join__type(graph: REVIEWS)
65
{
66
  body: String
67
  author: User @join__field(graph: REVIEWS, provides: "username")
68
  product: Product
69
}
70

71
type User
72
  @join__type(graph: REVIEWS, key: "id")
73
  @join__type(graph: USERS, key: "id")
74
{
75
  id: ID!
76
  username: String! @join__field(graph: REVIEWS, external: true) @join__field(graph: USERS)
77
  reviews: [Review] @join__field(graph: REVIEWS)
78
}

As you can see, the supergraph schema includes a lot of Federation-specific additions! These additions are used only by the router, and you'll never need to add them manually.

The router uses its supergraph schema to produce an API schema, which it exposes to clients as your actual GraphQL API. This schema cleanly and logically represents the combination of your subgraph schemas:

1
type Product {
2
  name: String!
3
  price: Int
4
  reviews: [Review]
5
  upc: String!
6
}
7

8
type Query {
9
  me: User
10
  topProducts(first: Int = 5): [Product]
11
}
12

13
type Review {
14
  author: User
15
  body: String
16
  product: Product
17
}
18

19
type User {
20
  id: ID!
21
  reviews: [Review]
22
  username: String!
23
}

Unlike the supergraph schema, this schema hides the fact that your GraphQL API is composed of multiple distinct GraphQL APIs.

JetBrains IDE Support

Composition