any pointers would be huge

👋 I recently read this excellent write up 

Lessor Known PostgreSQL Features

 https://hakibenita.com/postgresql-unknown-features#prevent-setting-the-value-of-an-auto-generated-key it was a top post on hacker news: https://news.ycombinator.com/item?id=29163319 One of the solid pieces of advice it has was to use an alternative primary key column type:

CREATE TABLE sale (
    id INT GENERATED ALWAYS AS IDENTITY PRIMARY KEY,
    sold_at TIMESTAMPTZ,
    amount INT
);

This blocks users from being able to write directly to the 

id

 field of a schema like above with an error:

db=# INSERT INTO sale (id, sold_at, amount) VALUES (2, now(), 1000);
ERROR:  cannot insert into column "id"
DETAIL:  Column "id" is an identity column defined as GENERATED ALWAYS.

The

@db.Int @default(autoincrement())

Generates the

SERIAL

column in postgres - is it possible to specify a custom column type?

model User {
  id   String  @id 'INT GENERATED ALWAYS AS IDENTITY PRIMARY KEY'
  test String?
}

Hey prisma guys! Does anyone has a working demo of prisma with latest Nextjs 12? In build I keep getting

Cant resolve

errors… - I use prisma in

getStaticProps

. Previously I had version 10.0.4 of Next and it worked. I use monorepo and Prisma is separate lib.

Hey Prisma Team! Is there a way to get freshly created ids when using

createMany

?

Is it expected/normal to see a 14MB increase in Vercel serverless function’s bundle size after adding Prisma? I’ve tried to only add

@prisma/client

and generate it using

npx prisma generate

to see if I could avoid adding

@prisma

but didn’t see any change.

I have a bug with Prisma. I'm getting:

Uncaught (in promise) Error:
Invalid `prisma.groupChatChannelGroup.findMany()` invocation:

  Value out of range for the type. The column `lastRead` contained an invalid datetime value with either day or month set to zero.

But the column doesn't exist in the table. I check out index.d.ts and I find this:

export type GroupChatChannelGroupByOutputType = {
    id: Buffer
    ...
    lastRead: Date
    ...
    _count: GroupChatChannelCountAggregateOutputType | null
    _avg: GroupChatChannelAvgAggregateOutputType | null
    _sum: GroupChatChannelSumAggregateOutputType | null
    _min: GroupChatChannelMinAggregateOutputType | null
    _max: GroupChatChannelMaxAggregateOutputType | null
  }

lastRead: Date shouldn't exist. How can I debug where this gets made? Thanks

Databases Access in Serverless Environments with the Prisma Data Proxy Databases connection management in serverless functions is a major issue for many developers. The Prisma Data Proxy (Early Access) solves this problem by managing a connection pool.

Hey all 👋 is there a way to conditionally do a

connect

? This doesn't seem to work:

location: {
          connect: existingLocation
            ? {
                id: existingLocation?.id,
              }
            : {},
          create: location,
        },

Hello guys, I have a doubt about the size of the Prisma Client inside a docker container in a microservice architecture, some of the microservices depends on Prisma, so the final image after building the docker image is getting more than 200mb only from the Prisma Client (node_modules/.prisma with 93mb and node_modules/@prisma with 115mb), do you guys know a better way to deal with this dependency of the Prisma Client in a microservice architecture?

I just have to say this .. Prisma is amazing!!! What's being created by the team is outstanding. The utility of Prisma is just off the chain. With MongoDB finally at a significant place in basically an unpaid advocate within the organizations that I contract. Thank you Team Prisma!!

Is there a relation between the number of query engines running to the connection pool setting or should the number of running query engine instances generally be fixed to a one?

When creating a record, is there now a way to include relations in the return value? Or would I have to make a separate query?

Hi all, I have an enum in my prisma schema like the below:

enum paddock_slope_enum {
  One__0____7__ @map("One (0° - 7°)")
  Two__8____15__ @map("Two (8° - 15°)")
  Three__16____25__ @map("Three (16° - 25°)")
  Four___26__ @map("Four (>26°)")
}

The generated enum becomes:

export const paddock_slope_enum: {
  One__0____7__: 'One__0____7__',
  Two__8____15__: 'Two__8____15__',
  Three__16____25__: 'Three__16____25__,
  Steep_Hill___26__: 'Steep_Hill___26__'
};

This is unexpected for me as I would have expected the values of the enum keys to be the map values in the schema? So my findMany GET API is returning the values with all the underscores - why is that? And can I get around it to get the value of the map?

Wondering if there's a straightforward way to delete a possibly orphaned record after disconnecting it? I have many-to-many user tagging and after disconnecting a tag from a user, I need to delete the tag if it's no longer referenced by other users. This is trivially easy with raw SQL but I couldn't find a way to do the delete with Prisma in one step:

prisma.$transaction([
      prisma.user.update({
        where: { id: input.userId },
        data: {
          tags: {
            disconnect: [{ id }],
          },
        },
      }),
      prisma.$executeRaw`DELETE FROM "Tag" WHERE id NOT IN (SELECT "B" FROM "_UserToTag")`,
    ])

Does anyone have any experience with writing unit tests in NestJS with Prisma?

I'm trying to create unit tests in NestJS with Prisma and found some examples and with the documentation on Prisma. Now trying to get it all working I'm running into some issues. context.ts

import { PrismaClient } from '.prisma/client';
import { mockDeep } from 'jest-mock-extended';
import { DeepMockProxy } from 'jest-mock-extended/lib/cjs/Mock';

export type Context = {
  prisma: PrismaClient;
};

export type MockContext = {
  prisma: DeepMockProxy<PrismaClient>;
};

export const createMockContext = (): MockContext => {
  return {
    prisma: mockDeep<PrismaClient>(),
  };
};

I'm getting the following error:

Type '{ $on: CalledWithMock<void, [eventType: "beforeExit", callback: (event: () => Promise<void>) => void]>; $connect: CalledWithMock<Promise<void>, []>; ... 13 more ...; readonly subMenu: DeepMockProxy<...>; } & PrismaClient<...>' is not assignable to type 'DeepMockProxy<PrismaClient<PrismaClientOptions, never, RejectOnNotFound | RejectPerOperation | undefined>>'.
      Type '{ $on: CalledWithMock<void, [eventType: "beforeExit", callback: (event: () => Promise<void>) => void]>; $connect: CalledWithMock<Promise<void>, []>; ... 13 more ...; readonly subMenu: DeepMockProxy<...>; } & PrismaClient<...>' is not assignable to type '{ $on: CalledWithMock<void, [eventType: "beforeExit", callback: (event: () => Promise<void>) => void]>; $connect: CalledWithMock<Promise<void>, []>; ... 13 more ...; readonly subMenu: DeepMockProxy<...>; }'.
        The types of '$on.calledWith' are incompatible between these types.
          Type '(...args: [eventType: "beforeExit", callback: (event: () => Promise<void>) => void] | [eventType: "beforeExit" | import("/Users/siemenblok/station21/node_modules/jest-mock-extended/lib/Matchers").Matcher<"beforeExit">, callback: ((event: () => Promise<...>) => void) | import("/Users/siemenblok/station21/node_modules...' is not assignable to type '(...args: [eventType: "beforeExit", callback: (event: () => Promise<void>) => void] | [eventType: "beforeExit" | import("/Users/siemenblok/station21/node_modules/jest-mock-extended/lib/cjs/Matchers").Matcher<"beforeExit">, callback: ((event: () => Promise<...>) => void) | import("/Users/siemenblok/station21/node_mod...'.
            Types of parameters 'args' and 'args' are incompatible.
              Type '[eventType: "beforeExit", callback: (event: () => Promise<void>) => void] | [eventType: "beforeExit" | import("/Users/siemenblok/station21/node_modules/jest-mock-extended/lib/cjs/Matchers").Matcher<"beforeExit">, callback: ((event: () => Promise<...>) => void) | import("/Users/siemenblok/station21/node_modules/jest-...' is not assignable to type '[eventType: "beforeExit", callback: (event: () => Promise<void>) => void] | [eventType: "beforeExit" | import("/Users/siemenblok/station21/node_modules/jest-mock-extended/lib/Matchers").Matcher<"beforeExit">, callback: ((event: () => Promise<...>) => void) | import("/Users/siemenblok/station21/node_modules/jest-mock...'.
                Type '[eventType: "beforeExit" | Matcher<"beforeExit">, callback: ((event: () => Promise<void>) => void) | Matcher<(event: () => Promise<void>) => void>]' is not assignable to type '[eventType: "beforeExit", callback: (event: () => Promise<void>) => void] | [eventType: "beforeExit" | Matcher<"beforeExit">, callback: ((event: () => Promise<void>) => void) | Matcher<(event: () => Promise<...>) => void>]'.
                  Type '[eventType: "beforeExit" | Matcher<"beforeExit">, callback: ((event: () => Promise<void>) => void) | Matcher<(event: () => Promise<void>) => void>]' is not assignable to type '[eventType: "beforeExit", callback: (event: () => Promise<void>) => void]'.
                    Type at position 0 in source is not compatible with type at position 0 in target.
                      Type '"beforeExit" | Matcher<"beforeExit">' is not assignable to type '"beforeExit"'.
                        Type 'Matcher<"beforeExit">' is not assignable to type '"beforeExit"'.

    15     prisma: mockDeep<PrismaClient>(),
           ~~~~~~

      context.ts:10:3
        10   prisma: DeepMockProxy<PrismaClient>;
             ~~~~~~
        The expected type comes from property 'prisma' which is declared here on type 'MockContext'

Does anyone know why? Or how I can fix this?

hey 🖐️

🎲 Care to cast your vote? 👀 We're very curious to see the results & your feedback! https://twitter.com/prisma/status/1459152858000113664

Does Prisma support filtering with IN by tuples, e.g.:

SELECT * FROM table
WHERE (letter, number) IN (('T', 7), ('R', 2))

where table is:

number | letter
  7    |   'T'
  2    |   'R'
  4    |   'T'

if not, joined workaround?

Question: Would it be possible to generate models using the Prisma Generator Specification? Looking to implement versioning for each data model defined in schema, and the idea was to write a generator that writes a

Version

model for each model. E.g.

model User {
  id: Int
  name: String
}

generated model:

model UserVersion {
  id: Int
  data: Json 
  userId Int
  user User @relation(fields: [userId], references: [id])
}

Greetings, and of course an associated question! Hurray, another question, just like you wanted! I'm trying to get Prisma1 to play along side Prisma2, and I have a Prisma2 schema that looks like this:

model User {
  id     String  @id @default(cuid()) @db.Char(30)
  email  String  @db.MediumText
  admin  Admin?
}
model Admin {
  id     String  @id @default(cuid()) @db.Char(30)
  name   String  @db.MediumText
  userId String  @db.Char(30)
  user   User    @relation(fields: [userId], references: [id], onUpdate: Restrict)
}

And in a Prisma1 GraphQL Playground I run this:

mutation ($data: UserCreateInput!) {
  createuser(data: $data) {
    id
    email
  }
}
variables
{
  "data" {
    "email": "<mailto:diety@supremebeing.org|diety@supremebeing.org>"
    "admin": {
      "create": {
        "name": "me"
      }
    }
  }
}

And I get an error about

(conn=3216) Field 'userId' doesn't have a default value

I also checked in MySQL and the foreign key constraint in

Admin

for

userId

references

id

in

User

Why isn't Prisma1 inserting

userId

into

Admin

when I create a user with a nested create for

Admin

? Is that something that Prisma1 can't do?

model User {
  id        Int       @id @default(autoincrement())
  createdAt DateTime  @default(now())
  email     String    @unique
  username  String    @db.VarChar(50)
  password  String
  ...
}

model Message {
  id        Int      @id @default(autoincrement())
  createdAt DateTime @default(now())
  subject   String
  body      String
}

I'm having trouble understanding how to best setup the relationships for a 1 on 1 private messaging system. I'd like for it to work similar to how Reddit handles private messages. Could someone point me in the right direction please.

when using planetscaledb, should we modify the connection_limit parameter since default is

num_physical_cpus * 2 + 1

but using planetscale we can use up to 10k concurrent requests

How would you query the last 10 recent posts in a database, get the post title, total user votes and if current id has voted? Schema looks like reddit, e.g. Users <-> Votes <-> Posts

Question on Prisma 2 middleware. The name of the model comes in params.model. How can I check if that model contains a specific property ? If we use the docs as an example, how could I check if language was actually defined as a property on 'Post' before setting it?

prisma.$use(async (params, next) => {
  if (params.model == 'Post' && params.action == 'create') {
    params.args.data.language = contextLanguage
  }

  return next(params)
})

Hi there! I was wondering if anyone knows of a good way to replicate this package or this implementation with NextJS, MySQL, and Prisma? Would Prisma be willing to expand on the Seed feature to allow for this kind of migration?

SDK question: Why does DMMF not include any data associated to defined

@@index

Writing a plugin that does versioning to store history of changes on a data model – for each

Datamodel

, would like to define a

DatamodelVersion

model which specifies some fields for storing history and some references to

Datamodel

In addition would like to add

versions: DatamodelVersion

to the original

Datamodel

schema. But in doing so, losing any

@@index

that is defined on the original

Datamodel

schema.

I just added a new model to existing data:

model Settings {
  id       String @id @default(cuid())
  timezone String @default("UTC")
  user     User[]
}

It threw migration error so I used

--create-only

but don't know how to fix it. I've been redirected to https://pris.ly/d/migrate-resolve but don't see anything there that helps me. The error is:

Applying migration `20211114104346_create_settings`
Error: P3018

A migration failed to apply. New migrations cannot be applied before the error is recovered from. Read more about how to resolve migration issues in a production database: <https://pris.ly/d/migrate-resolve>

Migration name: 20211114104346_create_settings

Database error code: 23502

Database error:
ERROR: column "settingsId" of relation "User" contains null values

DbError { severity: "ERROR", parsed_severity: Some(Error), code: SqlState("23502"), message: "column \"settingsId\" of relation \"User\" contains null values", detail: None, hint: None, position: None, where_: None, schema: Some("public"), table: Some("User"), column: Some("settingsId"), datatype: None, constraint: None, file: Some("tablecmds.c"), line: Some(5815), routine: Some("ATRewriteTable") }

How do I fix it?

Hello, I am wondering if anyone knows what the status of the Code-first Prisma schema AST/SDK project is?

I've just updated my project to 3.4.2 that uses sqlite and I'm getting timeout errors from the sqlclient. I posted a migration problem on stackoverflow (Connection error when seeding data with Prisma while running migration only - Stack Overflow), but am now realizing the timeout problem is happening seemingly randomly. After I got my app working and the data seeded, when I run my app, I'm getting timeout errors on simple things that don't make sense to me also. (return to soon...) Here is an error I get from a delete:

Invalid `prisma.noteOnTag.delete()` invocation:


  Error occurred during query execution:
ConnectorError(ConnectorError { user_facing_error: None, kind: ConnectionError(Timed out during query execution.) })

This is for a Pluralsight course I'm working on, and though the source is not public yet, I can share it if that helps.