is there any way to specify a connection timeout when fetching a prisma client? my aurora serverless DB might be asleep and i'd like to be able to wait a couple minutes for it to wake up

Hi, We are planning a multi-tenant app where each tenant will have their own database in different servers, we just want only change the database URL string and root the users to their db. Is this possible in Prisma? All the endpoints will be a lambda function.

Hi, this might be noob-question, but maybe you people can give me some guidance: I’m fetching a list of “groups” which includes different stuff like a name, type, other stuff, and then lastly a “date” (coming from a DateTime in the schema). Now, when I’m sending this response, the date is a

string

and not a

date

. I know I can create a

New Date()

on the frontend but my types are messing up because it’s set to

Date

-type in the schema but it’s actually a string in my code. What is the correct way to handle this?

Hello, Is there any to select everything but a field with prisma? If I want to select all fields on a table but not one, I need to set all the others to true Maybe I could just say something like exclude and tell which columns to exclude ?

Hello!

I am trying to use transactions in a typescript service. essentially I have an optional options object argument

options?: { $transaction?: Prisma.TransactionClient | null }

and the line

const prisma = options?.$transaction ?? this.prisma

(where

this.prisma

is the

PrismaClient

injected into my object) unfortunately Typescript seems to think

prisma

is always the

Prisma.TransactionClient

so I can't say

if (typeof prisma.$transaction !== 'undefined')

to narrow the scope between what should be

PrismaClient | Prisma.TransactionClient

Is there any better way to conditionally run a service method in a transaction? secondary question. Is there a way given a

Prisma.TransactionClient

that I can pass in Promises similar to how I can run

prisma.$transaction([prisma.model.action(), prisma.model.action2()])

? tertiary question, why does the

Prisma.TransactionClient

not have it's own $transaction method? At the very least it could just use the currently open transaction, at best it could optionally do a nested transaction or savepoint

Has anyone had this error?

Query createOneshipment_entry is required to return data, but found no record(s).

Reference (https://github.com/prisma/prisma/issues/12783)

With the new prisma queryRaw, if I am doing a

where id IN (${Prisma.join(ids)})

where the id is a uuid, I get the error

Raw query failed. Code: `42883`. Message: `db error: ERROR: operator does not exist: uuid = text

what do I need to do to cast those correctly?

Hello. could someone take a look at this? Something strange is happening with prisma and triggers. https://github.com/prisma/prisma/issues/14918

Hey, I started setting up my database and I decided I needed a table for users -

users

. In there, I have

id, email, password, createdAt

and so on. I soon realized that my users need roles as well, so, I created the

roles

column where I was to add an array of roles (I would have a

roles

table, connecting it all), but realized that MySQL doesn't(?) have support for arrays like that. I then settled on simply creating another table -

userMeta

, where I simply store

id, role

(where

id

is the user's id), in here, multiple rows can have the same user id, but not the same role, so, if I had an user with an id

1

and I had 2 roles -

administrator, user

, my

usersMeta

table would look as follows:

id  role
---------
1   administrator
1   user

In order for me to determine what roles an user has, I'd just simply run

SELECT role from user_meta WHERE id=1

. While this works, I just don't feel this is the right way to do things. I don't want to store roles as JSON, because then I can't query everything freely based on roles an user has or whatever else. Here are my current models:

enum Roles {
  ADMIN
  MODERATOR
  USER
}

model User {
  id       Int    @id @default(autoincrement())
  username String @unique
  email    String @unique

  firstName String @map("first_name")
  lastName  String @map("last_name")

  createdAt DateTime @default(now()) @map("created_at")
}

model UserMeta {
  id   Int
  role Roles @default(USER)

  @@id([id, role])
}

model Role {
  id           Int          @id
}

The database works, it's how I want it to be represented, I just don't know how to write it the "Prisma way". For example, right now, the

id

in

usersMeta

is any string. I'd like there to be a relation where that id can ONLY be an id from the

User

model.

Hi i've been racking my head on this for days. First here's the schema

// This is your Prisma schema file,
// learn more about it in the docs: <https://pris.ly/d/prisma-schema>

// The Models have been standardized this way to remove any confusion
// in_ : denotes that it is in these related field ex: mtg is in_boxes
// _id : is a field that requires the id of an other table like : user_id
// _rel: This is prisma generated field
datasource db {
  provider = "postgresql"
  url      = env("DATABASE_URL")
}

generator client {
  provider = "prisma-client-js"
}

model Games {
  id         Int      @id @default(autoincrement())
  createdAt  DateTime @default(now())
  updatedAt  DateTime @updatedAt
  short_hand String   @unique @db.VarChar(255)
  name       String   @unique @db.VarChar(255)
  in_boxes   Box[]
  in_sets    Sets[]
  vendor     String   @db.VarChar(255) @default("")
  sealed     String[] @db.VarChar(255)
}

model User {
  id         Int      @id @default(autoincrement())
  createdAt  DateTime @default(now())
  updatedAt  DateTime @updatedAt
  username   String   @unique @db.VarChar(255)
  password   String   @db.VarChar(255)
  role       Role     @default(USER)
  first_name String?  @db.VarChar(255)
  last_name  String?  @db.VarChar(255)
  stores     Store[]
}

model Store {
  id         Int      @id @default(autoincrement())
  createdAt  DateTime @default(now())
  updatedAt  DateTime @updatedAt
  name       String   @unique @db.VarChar(255)
  owner_id   Int?      
  users_list User[]
  boxes      Box[]
}

model Box {
  id                Int      @id @default(autoincrement())
  createdAt         DateTime @default(now())
  updatedAt         DateTime @updatedAt
  set               String   @db.VarChar(255)
  set_rel           Sets     @relation(fields: [set], references: [name])
  box_number        String   @db.VarChar(100)
  box_second_number String?  @db.VarChar(100)
  game_id           Int?     @default(1)
  game_rel          Games?   @relation(fields: [game_id], references: [id])
  store_id          Int?
  store_rel         Store?   @relation(fields: [store_id], references: [id])


}

model Sets {
  id        Int      @id @default(autoincrement())
  createdAt DateTime @default(now())
  updatedAt DateTime @updatedAt
  name      String   @unique @db.VarChar(255)
  code      String?  @db.VarChar(255)
  game_id   Int?     @default(1)
  game_rel  Games?   @relation(fields: [game_id], references: [id])
  edition   String?
  children  String[]
  in_boxes  Box[]


}

model Logs {
  id         Int      @id @default(autoincrement())
  createdAt  DateTime @default(now())
  updatedAt  DateTime @updatedAt
  store_id   Int
  user_id    Int      
  action     String   @db.VarChar(255)
  message    String   @db.VarChar(255)
}

enum Role {
  USER
  ADMIN
  CHRIS}

Then here's the query i'm trying to do:

const set = await context.prisma.sets.findFirst({
        where: {
          AND: [
            {
              OR: [
                {
                  name: {
                    contains: name,
                    mode: "insensitive",
                  },
                },
                {
                  code: {
                    contains: code,
                    mode: "insensitive",
                  },
                },
              ],
            },
            {
              in_boxes: {
               every :{
                store_id: {
                  equals: context.req.session.storeId,
                }
               }
              }
            }
          
          ],
          
        },
        select: {
          id: true,
          name: true,
          code: true,
          game_id: true,
          children: true,
          in_boxes: true,
        }
      });

I'm trying to return the set that contains either the name or the code but only return those set who are in the boxes who have that storeId (an Int) The first way I've tried ids by getting all the boxes and then _`set?.in_boxes.filter((box: any) => box.store_id === context.req.session.storeId);` however I soon realised that the more Set request I will get the more slower this will go_ The other way is that i tried ataching the in_boxes to the OR statements however it never returns the right set An other way i've though is to go with my box table to find the set via name or code however i think it would be too cumbersome

Hello everyone! I’m currently facing an issue, where I’m trying to filter on a postgres field of type

number

. I basically want to filter from an input field, kind of like a

contains

but for a number.. Is that something we can do by any chance? Thanks in advance 🙂

What is the suggested way to use read replicas with prisma? I see there are some proposed solutions with open issues.

Is it normal for the updatedAt column to always be updated even when no data is changed when using upsert: https://www.prisma.io/docs/concepts/components/prisma-client/relation-queries#update-or-create-a-related-record

const update = await prisma.post.update({
  where: {
    id: 6,
  },
  data: {
    author: {
      upsert: {
        create: {
          email: '<mailto:bob@prisma.io|bob@prisma.io>',
          name: 'Bob the New User',
        },
        update: {
          email: '<mailto:bob@prisma.io|bob@prisma.io>',
          name: 'Bob the existing user',
        },
      },
    },
  },
})

Do we represent json that is array first as

Json[]

type or

Json

and just shove in

[{key:value}, ...]

into it? Ideally i'd like to be able to do

jsonField.map((item) => ...)

in my code.

hi guys, qq. I'm running a migration inside an ECS container but i'm getting error:

Can't write to /application/node_modules/prisma please make sure you install "prisma" with the right permissions.

👀

package.json

dependencies: {
"@prisma/client": "^4.1.0",
},
devDepe {
   "prisma": "^4.1.0"
}

Any ideas? 😬 thanks in advance 🙂 The error occurs when running the command:

npx prisma generate --schema=./path/to/file/prisma/schema.prisma

in logs I can see:

Prisma schema loaded from ./path/to/file/prisma/schema.prisma
Error: Can't write to /application/node_modules/prisma please make sure you install "prisma" with the right permissions.

Hello, there was a problem during master deployment after changing from postgres url to prisma proxy url.

InvalidDatasourceError: Datasource "db" references an environment variable "POSTGRE_DATABASE_URL" that is not set

An error occurs. My env value is like this.

<prisma://XXX.prisma-data.com/?api_key=XXX>

Prisma Enabled Version:

4.2.1

Prisma.schema is as follows.

datasource db {

provider = "postgresql"

url = env("POSTGRE_DATABASE_URL")

}

Is there a reason why this error occurs?

Is there a possibility that Prisma will support Planetscale's Serverless Driver as an alternative to the Prisma Data Proxy?

what user does prisma use to generate a client on linux? meaning if you have

user     group      permissions   folder
myUser   nogroup    drwx-r-xr-x    src
myUser   nogroup    drwx-r-xr-x    node_modules/prisma
myUser   nogroup    drwx-r-xr-x    node_modules/.prisma
myUser   nogroup    drwx-r-xr-x    node_modules/@prisma

the api I have is executed by the user: `myUser`on startup, and it also runs a separated script as a child process to run migrations from, but i get the error:

Can't write to /application/node_modules/prisma please make sure you install "prisma" with the right permissions.

which makes me think I should add the

w

permission to prisma folders?

RUN chmod -R o+w /application/node_modules/.prisma /application/node_modules/@prisma /application/node_modules/prisma

the script i'm using to run migrations looks like:

const cmd1 = spawnSync('npx', ['prisma', 'generate', `--schema=${prismaSchema}`]);
    console.log('cmd1: ', cmd1.stdout.toString('utf8'), cmd1.stderr.toString('utf8'));

const cmd2 = spawnSync('npx', ['prisma', 'migrate', 'deploy', `--schema=${prismaSchema}`]);
    console.log('cmd2: ', cmd2.stdout.toString('utf8'), cmd2.stderr.toString('utf8'));

Could someone guide me here? 😅 All of these is running in a ECS container, any help is highly appreciated 🙂

I have a Disambiguating relations between two models like this

model Product {
  id          Int     @id @default(autoincrement())
  name        String?
  fees        Fee[]   @relation("fees")
  default_fee Fee?    @relation("default_fee")
}

model Fee {
  id             Int      @id @default(autoincrement())
  total          Float    @default(0)
  product_id     Int
  product        Product  @relation(name: "fees", fields: [product_id], references: [id], onDelete: Cascade)
  defaultable    Product? @relation(name: "default_fee", fields: [defaultable_id], references: [id])
  defaultable_id Int?     @unique
}

While fetching its works like expected. However while creating product with nested relations like this:

const product = await this.prisma.product.create({
      data: {
       ...data,
        default_fee: {
          create: {
            // asks for product id
            total: 500,
          },
        },
      },
    });

It asks me for product id. But it doesn't make sense to pass product_id as I wouldn't have the product id at that point. I know I can make it work by breaking the query down, but nested create doesnot seem to work for disambiguating relations. I'm curious wether it's issue on prisma or I'm missing something here Repo Link: https://github.com/sagarPakhrin/prisma-disambiguating-relations/blob/master/src/app.service.ts

any advice on how to run the migrations on the prod database when it's inside a VPC in GCP? I'd really really really like to avoid connecting to the VPC from GH actions to run that, there must be a better way, but upon google searching not a lot of 'meaty answers' have popped up so thought I'd ask here

Hey everyone- does someone knows if it's possible to directly connect to the DB from my Prisma 1 Cloud service? Looking to migrate all the data. Appreciate it

Hey guys could use some help, been at this all night with no luck, geatly appreciate it

model Domain {
  id                 String              @id @default(cuid())
  domain             String              @unique
  majesticDomainData MajesticDomainData?
}

model MajesticDomainData {
  id                     String                @id @default(cuid())
  domain                 Domain                @relation(fields: [domainId], references: [id])
  domainId               String                @unique
  topics                 MajesticTopicalData[]
}

model MajesticTopicalTrustFlow {
  id         String                @id @default(cuid())
  topic      String                @unique
  domainData MajesticTopicalData[]
}

model MajesticTopicalData {
  domainData         MajesticDomainData       @relation(fields: [domainDataId], references: [id])
  domainDataId       String
  topicalTrustFlow   MajesticTopicalTrustFlow @relation(fields: [topicalTrustFlowId], references: [id])
  topicalTrustFlowId String
  topicalPercent     Decimal                  @db.Decimal(5, 4)

  @@id([domainDataId, topicalTrustFlowId])
}

prisma.domain.update({
                            where: {<http://asd.com|asd.com>},
                            data: {
                                majesticDomainData: {
                                    create: {
                                        topics: {
                                            create: [
                                                {
                                           topic: 'health',
                                           topicalPercent: 0.24
                                                },
                                                {
                                          topic: 'rec',
                                         topicalPercent: 0.33
                                                },
                                            ]
                                        },
                                    }
                                }
                            },
                        })

+             topicalTrustFlow: {
+               create?: MajesticTopicalTrustFlowCreateWithoutDomainInput | MajesticTopicalTrustFlowUncheckedCreateWithoutDomainInput,
+               connectOrCreate?: MajesticTopicalTrustFlowCreateOrConnectWithoutDomainInput,
+               connect?: MajesticTopicalTrustFlowWhereUniqueInput
+             }

Im not sure if im connecting the many to many relationship correctly, its asking for my explicit table

Hi guys, is there a way I can run seeds from several locations on command? In my project I have 2 packages each with its own prisma folder and everything in it, including seeds. the suggestions from the doc doesn't work for me as I need to specify 2 different path for seeds and prior I need to dynamically setup the

DATABASE_URL

which is working fine for migrations, but now for seeding i'm unsure how to proceed 🤔 In my script I tried running:

const seeds = spawnSync('npx', ['ts-node', './src/package/package_1/prisma/seed.ts']);

but i get the error:

node:internal/modules/cjs/loader:936
  throw err;
  ^

Error: Cannot find module './seed.ts'
Require stack:
- /application/src/packages/package_1/prisma/imaginaryUncacheableRequireResolveScript
    at Function.Module._resolveFilename (node:internal/modules/cjs/loader:933:15)
    at Function.resolve (node:internal/modules/cjs/helpers:108:19)
    at requireResolveNonCached (/application/node_modules/ts-node/dist/bin.js:549:16)
    at getProjectSearchDir (/application/node_modules/ts-node/dist/bin.js:519:40)
    at phase3 (/application/node_modules/ts-node/dist/bin.js:267:27)
    at bootstrap (/application/node_modules/ts-node/dist/bin.js:47:30)
    at main (/application/node_modules/ts-node/dist/bin.js:33:12)
    at Object.<anonymous> (/application/node_modules/ts-node/dist/bin.js:579:5)
    at Module._compile (node:internal/modules/cjs/loader:1105:14)
    at Object.Module._extensions..js (node:internal/modules/cjs/loader:1159:10) {
  code: 'MODULE_NOT_FOUND',
  requireStack: [
    '/application/src/packages/package_1/prisma/imaginaryUncacheableRequireResolveScript'
  ]
}

Any ideas? 😄 Thanks in advance 🙂

Hey Mischa 👋 Could you please elaborate? Are you talking about

@prisma

folder in node_modules?

Hey all, sorry if I missed the v3 to v4 migration guide, but does someone have a link for specifically dealing with this? I have quite a generic prisma app and wondering how tricky it will be to move … have we got wholescale breaking changes, or just a small amount of breaking changes? Thanks 🙂

how can I use

prisma.user.updateMany

and disconnect a list using

data: {discoonect: {id: 1}}

?

Hey there everyone: We have a development environment with a remote development postgres server (hence, everything which is merged into DEV trunc is running against remote PG development server). I've read the following: https://www.prisma.io/docs/guides/database/developing-with-prisma-migrate/team-development The document mentions: "Ania and Javier make additive changes to the schema in their local environment and generate migrations" Does this then recommend a developer run a shadow DB instance (local PG db) against the remote postgress instance to obtain the latest changes from the development environment in order to test/develop against? I'm trying to understand the typical prisma development workflow using a PG Appreciate anyone's feedback here 👍

This message was deleted.