alright upon further poking around. it appears that two of my relations are returning undefined.

i wrote a script and checked every row + relation. every relation/ID exists in my entire set. when I loop through each row, there are zero nulls

it appears that when I request to many rows it starts to fail

hmm I found the issue. one of the IDs don't exist. 🤦 alright. how can I enforce relations, check that the IDs exist. I enabled referential Integrity

generator client {
  provider        = "prisma-client-js"
  previewFeatures = ["dataProxy", "referentialIntegrity"]
}

datasource db {
  provider             = "mysql"
  url                  = env("DATABASE_URL")
  referentialIntegrity = "prisma"
}

Hi there, I am total newbie in data model. Can I have your advice here? I am trying to build an online order app using nextjs and planetscale. Basically, user can create digital menu and share it so customer can order via whatsapp. Attached is the schema I have. Features: 1. user has many restaurants ( menu ) 2. menu has many sections 3. menu has many items 4. item has many options / variants 5. user has many orders 6. order has many order items for variants and options, user can create it and assign to any items. Or maybe just assign to a section, I am not sure which works better...

generator client {
  provider        = "prisma-client-js"
  previewFeatures = ["referentialIntegrity"]
}

datasource db {
  provider             = "mysql"
  url                  = env("DATABASE_URL")
  referentialIntegrity = "prisma"
}

// user has many restaurants
model User {
  id          String       @id @unique @default(cuid())
  name        String
  restaurants Restaurant[]
  orders      Order[]
}

// restaurant has many sections, items

model Restaurant {
  id     String @id @unique @default(cuid())
  name   String
  userId String

  User              User               @relation(fields: [userId], references: [id])
  sections          Section[]
  items             Item[]
  itemOptions       ItemOption[]
  itemOptionValues  ItemOptionValue[]
  itemVariants      ItemVariant[]
  itemVariantValues ItemVariantValue[]
  orders            Order[]
  address           Address?

}

model Section {
  id           String  @id @unique @default(cuid())
  name         String
  restaurantId String?

  Restaurant Restaurant? @relation(fields: [restaurantId], references: [id])
  items      Item[]
}

model Item {
  id           String  @id @unique @default(cuid())
  name         String
  price        Decimal
  restaurantId String?
  sectionId    String?

  Restaurant   Restaurant?   @relation(fields: [restaurantId], references: [id])
  Section      Section?      @relation(fields: [sectionId], references: [id])
  itemOptions  ItemOption[]
  itemVariants ItemVariant[]
}

// multiple choices eg: extra cheese, extra sauce etc.
model ItemOption {
  id           String  @id @unique @default(cuid())
  name         String
  restaurantId String?
  itemId       String?

  Restaurant Restaurant? @relation(fields: [restaurantId], references: [id])
  Item       Item?       @relation(fields: [itemId], references: [id])
}

model ItemOptionValue {
  id           String  @id @unique @default(cuid())
  name         String
  price        Decimal
  restaurantId String?

  Restaurant Restaurant? @relation(fields: [restaurantId], references: [id])
}

// sizes, flavor and toppings etc
model ItemVariant {
  id           String  @id @unique @default(cuid())
  name         String
  restaurantId String?
  itemId       String?

  Restaurant Restaurant? @relation(fields: [restaurantId], references: [id])
  Item       Item?       @relation(fields: [itemId], references: [id])
}

model ItemVariantValue {
  id           String  @id @unique @default(cuid())
  name         String
  price        Decimal
  restaurantId String?

  Restaurant Restaurant? @relation(fields: [restaurantId], references: [id])
}

model Order {
  id           String  @id @unique @default(cuid())
  name         String
  total        Decimal
  userId       String?
  restaurantId String?

  orderItems OrderItem[]
  User       User?       @relation(fields: [userId], references: [id])
  Restaurant Restaurant? @relation(fields: [restaurantId], references: [id])
}

model OrderItem {
  id       String  @id @unique @default(cuid())
  name     String
  quantity Int
  price    Decimal
  orderId  String?

  Order Order? @relation(fields: [orderId], references: [id])
}

model Address {
  id           String  @id @unique @default(cuid())
  name         String
  latitudes    Decimal @db.Decimal(8, 6)
  longitudes   Decimal @db.Decimal(9, 6)
  restaurantId String  @unique

  Restaurant Restaurant @relation(fields: [restaurantId], references: [id])
}

Am I doing it correctly? I wanted to be able to list the orders, number of an item being ordered for the owner as a stats/report. Also, an order summary for the customer to refer to as a receipt. Thanks for your help!

Hello everyone, can anyone explain how i get the properties of the object that you get back from the json type of prisma? It says that the value .name not exists. I'm in typescript

Hi all, I'm working on a feature where it's possible many resources can be created in a single endpoint. I've tried using

createMany

as the

skipDuplicates

flag is a useful feature for me, plus

createMany

is more performant than a loop with multiple single

create

. My issue is that

createMany

does not return the created values, and I need those values to perform another action - e.g. a notification email. I have found this thread, but nothing in there really helps my case https://github.com/prisma/prisma/issues/8131. I thought about using a loop and storing the promises in an array, then waiting for them all to be settled etc, however I haven't been able to get this to work and Prisma/TS isn't happy. Any suggestions greatly appreciated!

👋 Hello Everyone, This is my first post, I need to perform Prisma query example:

SELECT * FROM employee_profile WHERE month(date_of_birth) = 12 AND year(date_of_birth) = 1990

I searched for the Prisma docs, but I couldn't find the equivalent Prisma functions to use the SQL functions like

month()

,

year()

Is there any solution ?

Hello everyone

does anyone have experience with prisma and cloudflare? I run into an error where it states that I have used over 50 subrequests from cloudflare. i wonder if its related to the proxy...

Is there a possible way to extend the 'JsonArray' class? It gives me a error whenever i try to extend it

Any way to query

created < NOW() - INTERVAL '1 day'

?

Hey guys! Is there no way to change the type/interface of a Json field?

Is there a way to insert the Date.now() into a mysql DB?

As like, what datatype should it be

Hi all , please how do I go about this. When a field is

DateTime @db.Time()

Should I pass in a JavaScript date object or just a time string. Will the db.Time() format it for me if I were to pass in a date object?

Hi all, I’m a long time (2+ years) user of Prisma1, but I’m new to the slack. I just upgraded to Prisma2 this weekend (it wasn’t as painful as I thought it would be). I’m liking the new features (e.g. transactions) and syntax. However, one of the things I’m really missing is the auto-generated “WhereInputs” that Prisma1 gave you. It was very handy to be able to create a little query with custom where filtering specific to that query (often a one-off on the frontend) and not have to write a new backend service. I understand not wanting to expose the entire CRUD, but having the ability to generate all the WhereInputs for my schema for querying on the frontend would be helpful (I really don’t want to do it by hand). Does this exist? Has anyone else had this need? Thanks, Bailey

Hi folks! I am building a backend for football/soccer scouting app. Among many others, I need

clubs

,

competitions (leagues)

and

seasons

tables in my database. I'm also using a table called

competition_participations

to keep track of in which league was the club participating during particular season. I'm kinda struggling with modeling competitions (leagues) data. My problem is that there are leagues that are divided into groups and the club is participating in that particular group, but there are also top-level leagues that are not divided into any groups. Let's talk about Bundesliga here, as Prisma's headquarters is in Berlin 😄 Here is an example: 1. Bayern participates in

Bundesliga

, no groups 2. Heidenheim participates in

2. Bundesliga

, no groups 3. FC Kaiserslautern participates in

3. Liga

, no groups 4. VfB Oldenburg participates in

Regionalliga North

- top level competition is

Regionalliga

, group is

North

5. Alemannia Aachen participates in

Regionalliga West

- top level competition is

Regionalliga

, group is

West

How would you model that? I was thinking of something like

model Division {
  id          String   @id @default(cuid())
  name        String
  level.      Int // 1, 2, 3, etc..
  isJunior    Boolean @default(false)
  isWomen     Boolean @default(false)
  createdAt   DateTime @default(now())
  updatedAt   DateTime @updatedAt

  // Relation fields
  country   Country                    @relation(fields: [countryId], references: [id])
  countryId String
  league League[]
}

model League {
  id            String      @id @default(cuid())
  name          String

  // Relation fields
  division   Division @relation(fields: [divisionId], references: [id])
  divisionId String
}

But it seems a little bit dirty to have to create

Bundesliga

as a division and as a league. Do you have any recommendations?

Hello, Currently struggling to create what is probably a pretty straight forward query, but here's my question: Currently, I have 2 tables, 

publication

 and 

publicationStatus

. A publication will looking something like:

{
  "id": "ckyil950d00027v2str5ljo7h",
  "url_slug": "ckyil950e00037v2s4mqxvaho",
  "type": "PEER_REVIEW",
  "title": "Intersting review",
  "content": "Content is optional at this stage",
  "doi": "1093/ajae/aaq063",
  "createdBy": "test-user-1",
  "createdAt": "2022-01-17T11:12:50.845Z",
  "updatedAt": "2022-01-17T11:12:50.847Z",
  "publicationStatus": [
    {
      "status": "LIVE",
      "createdAt": "2022-01-19T11:12:50.846Z",
      "id": "ckyil950e00047v2sx4urbfte"
    },
    {
      "status": "DRAFT",
      "createdAt": "2022-01-17T11:12:50.846Z",
      "id": "ckyil950e00047v2sx4urbfth"
    }
  ],
  "user": {
    "id": "test-user-1",
    "firstName": "Test",
    "lastName": "User 1"
  }
}

Where 

publication

 has a 1 to many relationship with 

publicationStatus

. What I need to do is a 

find

 query where it only returns a 

publication

 if the latest 

publicationStatus

 for that 

publication

, has a 

status

 of 

LIVE

. The closest I could come to is this psuedo code (does not work, but demonstrates a picture of what I am trying to do):

await prisma.publication.findFirst({
    where: {
        id,
        publicationStatus: {
            where: {
                status: 'LIVE'
            },
            take: 1,
            orderBy: {
                createdAt: 'desc'
            }
        },
    }
});

What is the easiest way to check if a PrismaClient instance is closed/disconnected?

Any prisma maintainers maybe having knowlegde of any secret attributes for this or something? 😄

Hi all , please can someone help with this . I am getting a

PrismaClientKnownRequestError: 
Invalid prisma.restaurant.create() invocation:

Failed to validate the query: Unable to match input value to any allowed input type for the field. Parse errors: [Query parsing/validation error at

Yet my data is being created in the database and it still returns this error. Any help is greatly appreciated.

Is there a way to automatically generate a Graql Schema from the Prisma Schema?

Hi all, I am really having trouble using prisma studio with a larger data base, because prisma studio handles n:m relations in a way that makes it incredible slow. Some of these relations have over 10000 connections. I guess Prisma studio uses the include option for all attributes. This leads to more than 10,000 connections being requested and sent. Since this leads to very large amount of data, it results in a timeout/freeze. For example, you have a model

user

with the attribute

follower

. A user has 10k followers. Prisma studio would then request the data of all 10k followers as soon as this specific user is requested. Even if you unselect certain attributes in Prisma studio, they are still requested. 1. does anyone else has this kind of trouble and already found a solution/workaround? 2. why does prisma studio use include to show the number of relations instead of _count? This would probably speed up the whole process by a lot, I guess.

Hi I am using Docker Compose to run my database engine. And now I face this Error. What is it?

$ npx prisma migrate dev
Environment variables loaded from .env
Prisma schema loaded from prisma/schema.prisma
Datasource "db": PostgreSQL database "nest_base", schema "public" at "localhost:5432"

Already in sync, no schema change or pending migration was found.

Error: ENOENT: no such file or directory, copyfile '/path/to/node_modules/@prisma/index.d.ts' -> '/path/to/node_modules/@prisma/client/index.d.ts'

Hey, I try to get all assets/items and count their downloads just of the last 7 days. Counting all downloads, or getting the full download model back (without count aggregate, which I dont want) works. Is it just not possible yet, to filter aggregates? I really don't know how to make this work. Any help would be much appreciated!

What I basically need:

Whenever i try to use the .delete() function it gives me a error.. My code is

await prisma.tickets.delete({
        where: {
            channelId: interaction.channelId,
            guildId: interaction.guildId
        }
    });

And my error is

Type '{ channelId: any; guildId: any; }' is not assignable to type 'ticketsWhereUniqueInput'.
  Object literal may only specify known properties, and 'channelId' does not exist in type 'ticketsWhereUniqueInput'.

hey everyone! 👋 My backend is receiving an array of objects to update in the database. At the moment, to achieve this I am having to loop through the objects and create unique requests and I am then executing all these individual prisma queries within a transaction. Is there a better way of doing it? When i look at the generated SQL there are a large number of queries performed individually vs in aggregate

Hi, is there a way to reference an existing table that *is not in Prisma'*s schema? for example i have existing User table. Im creating a Post model in Prisma that i want to have a foreign key to User. If there is not built-in way, the best option is to change the generated migration files that responsible on the addition of Post model?