This example shows how to implement a GraphQL server with TypeScript with the following stack:
- Apollo Server: HTTP server for GraphQL APIs
- GraphQL Nexus: GraphQL schema definition and resolver implementation
- Prisma Client: Databases access (ORM)
- Prisma Migrate: Database migrations
- SQLite: Local, file-based SQL database
- Getting Started
- Using the GraphQL API
- Evolving the app
- Switch to another database (e.g. PostgreSQL, MySQL, SQL Server)
- Next steps
Download this example:
curl https://codeload.github.com/prisma/prisma-examples/tar.gz/latest | tar -xz --strip=2 prisma-examples-latest/typescript/graphql
Install npm dependencies:
cd graphql
npm install
Alternative: Clone the entire repo
Clone this repository:
git clone [email protected]:prisma/prisma-examples.git --depth=1
Install npm dependencies:
cd prisma-examples/typescript/graphql
npm install
Run the following command to create your SQLite database file. This also creates the User
and Post
tables that are defined in prisma/schema.prisma
:
npx prisma migrate dev --name init
When npx prisma migrate dev
is executed against a newly created database, seeding is also triggered. The seed file in prisma/seed.ts
will be executed and your database will be populated with the sample data.
Launch your GraphQL server with this command:
npm run dev
Navigate to http://localhost:4000 in your browser to explore the API of your GraphQL server in a GraphQL Playground.
The schema that specifies the API operations of your GraphQL server is defined in ./schema.graphql
. Below are a number of operations that you can send to the API using the GraphQL Playground.
Feel free to adjust any operation by adding or removing fields. The GraphQL Playground helps you with its auto-completion and query validation features.
query {
feed {
id
title
content
published
author {
id
name
email
}
}
}
See more API operations
{
draftsByUser(
userUniqueInput: {
email: "[email protected]"
}
) {
id
title
content
published
author {
id
name
email
}
}
}
mutation {
signupUser(data: { name: "Sarah", email: "[email protected]" }) {
id
}
}
mutation {
createDraft(
data: { title: "Join the Prisma Slack", content: "https://slack.prisma.io" }
authorEmail: "[email protected]"
) {
id
viewCount
published
author {
id
name
}
}
}
mutation {
togglePublishPost(id: __POST_ID__) {
id
published
}
}
Note that you need to replace the __POST_ID__
placeholder with an actual id
from a Post
record in the database, e.g.5
:
mutation {
togglePublishPost(id: 5) {
id
published
}
}
mutation {
incrementPostViewCount(id: __POST_ID__) {
id
viewCount
}
}
Note that you need to replace the __POST_ID__
placeholder with an actual id
from a Post
record in the database, e.g.5
:
mutation {
incrementPostViewCount(id: 5) {
id
viewCount
}
}
{
feed(
searchString: "prisma"
) {
id
title
content
published
}
}
{
feed(
skip: 2
take: 2
orderBy: { updatedAt: desc }
) {
id
updatedAt
title
content
published
}
}
{
postById(id: __POST_ID__ ) {
id
title
content
published
}
}
Note that you need to replace the __POST_ID__
placeholder with an actual id
from a Post
record in the database, e.g.5
:
{
postById(id: 5 ) {
id
title
content
published
}
}
mutation {
deletePost(id: __POST_ID__) {
id
}
}
Note that you need to replace the __POST_ID__
placeholder with an actual id
from a Post
record in the database, e.g.5
:
mutation {
deletePost(id: 5) {
id
}
}
Evolving the application typically requires two steps:
- Migrate your database using Prisma Migrate
- Update your application code
For the following example scenario, assume you want to add a "profile" feature to the app where users can create a profile and write a short bio about themselves.
The first step is to add a new table, e.g. called Profile
, to the database. You can do this by adding a new model to your Prisma schema file file and then running a migration afterwards:
// ./prisma/schema.prisma
model User {
id Int @default(autoincrement()) @id
name String?
email String @unique
posts Post[]
+ profile Profile?
}
model Post {
id Int @id @default(autoincrement())
createdAt DateTime @default(now())
updatedAt DateTime @updatedAt
title String
content String?
published Boolean @default(false)
viewCount Int @default(0)
author User? @relation(fields: [authorId], references: [id])
authorId Int?
}
+model Profile {
+ id Int @default(autoincrement()) @id
+ bio String?
+ user User @relation(fields: [userId], references: [id])
+ userId Int @unique
+}
Once you've updated your data model, you can execute the changes against your database with the following command:
npx prisma migrate dev --name add-profile
This adds another migration to the prisma/migrations
directory and creates the new Profile
table in the database.
You can now use your PrismaClient
instance to perform operations against the new Profile
table. Those operations can be used to implement queries and mutations in the GraphQL API.
First, add a new GraphQL type via Nexus' objectType
function:
// ./src/schema.ts
+const Profile = objectType({
+ name: 'Profile',
+ definition(t) {
+ t.nonNull.int('id')
+ t.string('bio')
+ t.field('user', {
+ type: 'User',
+ resolve: (parent, _, context) => {
+ return context.prisma.profile
+ .findUnique({
+ where: { id: parent.id || undefined },
+ })
+ .user()
+ },
+ })
+ },
+})
const User = objectType({
name: 'User',
definition(t) {
t.nonNull.int('id')
t.string('name')
t.nonNull.string('email')
t.nonNull.list.nonNull.field('posts', {
type: 'Post',
resolve: (parent, _, context) => {
return context.prisma.user
.findUnique({
where: { id: parent.id || undefined },
})
.posts()
},
+ t.field('profile', {
+ type: 'Profile',
+ resolve: (parent, _, context) => {
+ return context.prisma.user.findUnique({
+ where: { id: parent.id }
+ }).profile()
+ }
+ })
},
})
Don't forget to include the new type in the types
array that's passed to makeSchema
:
export const schema = makeSchema({
types: [
Query,
Mutation,
Post,
User,
+ Profile,
UserUniqueInput,
UserCreateInput,
PostCreateInput,
PostOrderBy,
DateTime,
],
// ... as before
}
Note that in order to resolve any type errors, your development server needs to be running so that the Nexus types can be generated. If it's not running, you can start it with npm run dev
.
// ./src/schema.ts
const Mutation = objectType({
name: 'Mutation',
definition(t) {
// other mutations
+ t.field('addProfileForUser', {
+ type: 'Profile',
+ args: {
+ userUniqueInput: nonNull(
+ arg({
+ type: 'UserUniqueInput',
+ }),
+ ),
+ bio: stringArg()
+ },
+ resolve: async (_, args, context) => {
+ return context.prisma.profile.create({
+ data: {
+ bio: args.bio,
+ user: {
+ connect: {
+ id: args.userUniqueInput.id || undefined,
+ email: args.userUniqueInput.email || undefined,
+ }
+ }
+ }
+ })
+ }
+ })
}
})
Finally, you can test the new mutation like this:
mutation {
addProfileForUser(
userUniqueInput: {
email: "[email protected]"
}
bio: "I like turtles"
) {
id
bio
user {
id
name
}
}
}
Expand to view more sample Prisma Client queries on Profile
Here are some more sample Prisma Client queries on the new Profile
model:
const profile = await prisma.profile.create({
data: {
bio: 'Hello World',
user: {
connect: { email: '[email protected]' },
},
},
})
const user = await prisma.user.create({
data: {
email: '[email protected]',
name: 'John',
profile: {
create: {
bio: 'Hello World',
},
},
},
})
const userWithUpdatedProfile = await prisma.user.update({
where: { email: '[email protected]' },
data: {
profile: {
update: {
bio: 'Hello Friends',
},
},
},
})
If you want to try this example with another database than SQLite, you can adjust the the database connection in prisma/schema.prisma
by reconfiguring the datasource
block.
Learn more about the different connection configurations in the docs.
Expand for an overview of example configurations with different databases
For PostgreSQL, the connection URL has the following structure:
datasource db {
provider = "postgresql"
url = "postgresql://USER:PASSWORD@HOST:PORT/DATABASE?schema=SCHEMA"
}
Here is an example connection string with a local PostgreSQL database:
datasource db {
provider = "postgresql"
url = "postgresql://janedoe:mypassword@localhost:5432/notesapi?schema=public"
}
For MySQL, the connection URL has the following structure:
datasource db {
provider = "mysql"
url = "mysql://USER:PASSWORD@HOST:PORT/DATABASE"
}
Here is an example connection string with a local MySQL database:
datasource db {
provider = "mysql"
url = "mysql://janedoe:mypassword@localhost:3306/notesapi"
}
Here is an example connection string with a local Microsoft SQL Server database:
datasource db {
provider = "sqlserver"
url = "sqlserver://localhost:1433;initial catalog=sample;user=sa;password=mypassword;"
}
Here is an example connection string with a local MongoDB database:
datasource db {
provider = "mongodb"
url = "mongodb://USERNAME:PASSWORD@HOST/DATABASE?authSource=admin&retryWrites=true&w=majority"
}
Because MongoDB is currently in Preview, you need to specify the previewFeatures
on your generator
block:
generator client {
provider = "prisma-client-js"
previewFeatures = ["mongodb"]
}
- Check out the Prisma docs
- Share your feedback in the
prisma2
channel on the Prisma Slack - Create issues and ask questions on GitHub
- Watch our biweekly "What's new in Prisma" livestreams on Youtube