This is a mock of the Prisma API intended for unit testing. All the data is stored in memory.
The library jest-mock-extended
is used, which means that if functionality you need is not implemented yet, you can mock it yourself.
Simple example how to create a prisma mock instance:
import createPrismaMock from "prisma-mock"
let client
beforeEach(() => {
client = createPrismaMock()
})
An example how to mock a global prisma instance, as the default export in a "db" directory (like blitzjs):
import createPrismaMock from "prisma-mock"
import { mockDeep, mockReset } from "jest-mock-extended"
jest.mock("db", () => ({
__esModule: true,
...jest.requireActual("db"),
default: mockDeep(),
}))
import db, { Prisma } from "db"
beforeEach(() => {
mockReset(db)
createPrismaMock({}, Prisma.dmmf.datamodel)
})
createPrismaMock(
data: PrismaMockData<P> = {},
datamodel?: Prisma.DMMF.Datamodel,
client = mockDeep<P>(),
options: {
caseInsensitive?: boolean
} = {}
): Promise<P>
You can optionally start up a pre-filled db, by passing in an object containing keys for tables, and values as arrays of objects (though using create
is preferred). Example:
createPrismaMock({
user: [
{
id: 1,
name: "John Doe",
accountId: 1,
},
],
account: [
{
id: 1,
name: "Company",
},
],
})
The datamodel of the prisma client, value of Prisma.dmmf.datamodel
.
A jest-mock-extended
instance. If not provided, a new instance is created.
If true, all string comparisons are case insensitive.
Most common cases are covered, but not everything. Here is a rough list of the supported features:
- findUnique,
- findUniqueOrThrow,
- findMany,
- findFirst,
- findFirstOrThrow,
- create,
- createMany
- delete,
- update,
- deleteMany,
- updateMany
- upsert
- count
- TODO: aggregate
- TODO: groupBy
- distinct
- include
- where
- select
- orderBy
- TODO: select: _count
- create
- createMany
- update
- updateMany
- delete
- deleteMany
- connect
- disconnect
- TODO: set
- TODO: connectOrCreate
- TODO: upsert
- equals
- gt
- gte
- lt
- lte
- not
- in
- notIn
- contains
- startWith
- endsWith
- AND
- OR
- NOT
- mode
- TODO: search
- some
- every
- none
- TODO: is
TODO (set, push)
TODO (has, hasEvery, hasSome, isEmpty, equals)
- increment
- decrement
- multiply
- divide
- set
TODO (path, string_contains, string_starts_with, string_ends_with, array_contains, array_starts_with, array_ends_with)
- @@id
- @default
- @unique (TODO: no error if duplicate)
- @@unique (TODO: no error if duplicate)
- @relation
- @updatedAt: Partially supported, value is set at creation (TODO: update value on update)
- autoincrement()
- TODO: auto()
- cuid()
- TODO: uuid()
- now()
- TODO: dbgenerated()
- onDelete (SetNull, Cascade)
- TODO: onDelete: Restrict, NoAction, SetDefault
- TODO: onUpdate
- $transaction
- $transaction (interactive)
- TODO: $transaction (isolation)
Create a .env-cmdrc
file in the root of your project with the following content:
{
"postgres": {
"PROVIDER": "postgresql",
"DATABASE_URL": "postgresql://postgres:postgres@localhost:5432/postgres?schema=public"
}
}
Create your tests in the __tests__
directory. You can use snapshot testing with either expect(res).toMatchSnapshot()
or expect(res).toMatchInlineSnapshot()
. This captures the result of your tests in a snapshot, which you can use to compare agains prisma-mock results.
Note: If you choose to use snapshot testing, make shore to first run your tests against the real database to create a snapshot of the expected result.
To run tests against a postgres database, run the following command:
yarn run test:postgres
To run tests against prisma-mock (in-memory database), run:
yarn test