looks like the above is a WSL2 only problem

Any ideas how to make seeding work properly with a node server with “type”: “module”? When I change the seed file to seed.cjs, prisma doesn’t find it.

Does anybody know how to provide the

binaryTargets

value as an environment variable? I’m deploying to AWS and setting the config to

binaryTargets = ["native", "rhel-openssl-1.0.x"]

works just fine, however I don’t want to include the

native

binary as it increases the lambda size considerably

Hi, I’m trying to use

createMany

, but it seems it couldn’t be found.

Property 'createMany' does not exist on type.

I’m using

v2.26.0

and sqlite. I’ve tried to put it into previewFeature, but I then get

Please either remove the "createMany" feature flag or use any other database type that Prisma supports: postgres, mysql or sqlserver.

Any idea?

for an implicit many-many relationship, when connecting/disconnecting, does it remove the record from the join table? or how does it really behave?

I’ve also updated our Prisma version from

2.6.2

to

2.26.0

and get now query times that are twice as slow. From 56ms to 121ms. Still sqlite. Any idea why there is a performance degradation between those versions?

Can anybody help me? I want search with similarity, unaccent and lower, I can do it with this query (postgresql)

SELECT
	*
FROM
	public."City"
WHERE
	word_similarity(LOWER(unaccent(name)), LOWER(unaccent('mexico'))) > 0.3
ORDER BY
	name

how would I do this with prisma? Note: I already have the plugins installed in the migrations

Is it possible to set a default for an array i.e.

roles     Role[]     @default([Role.USER])

? I'm getting a syntax error.

If anybody has any idea on how to set this properly, please let me know

hi !! our team uses the nexus-prisma-plugin and mainly uses the t.model and t.crud. we know that the prisma teamp is developing nexus-prisma, so what i’m wondering is.. 1. in nexus-prisma github page, t.model and t.crud status is long term. will it take long time to use this feature ? 2. if implemented, will it have the same functionality as the current one ? thank you for your effort 👍

hey!

how do you reset one table? I am familiar with

npx prisma migrate reset

but not for dropping a single relation

When are Geo data types expected in Prisma?

Hello team, I’m using Prisma@2.26 at the moment and I’m trying to set an environment variable for

client.binaryTargets

. It looks like there’s this issue which has been addressed in this PR. However, I’m still having problems, and may be someone can help clarify how I’m supposed to use it. Here’s the related part in my schema:

generator client {
  provider      = "prisma-client-js"
  binaryTargets = env("PRISMA_BINARY_TARGET") // I also tried to use `[env("PRISMA_BINARY_TARGET")]` but getting the same errors regardless
}

I have

JSON.parse

error when I try to do this:

PRISMA_BINARY_TARGET=native prisma generate

. Note that this was working in version 2.22 I think it’s because the parse function expects the env to be a JSON string 🤔 Anyways, running the following command works:

PRISMA_BINARY_TARGET='[\"native\"]' prisma generate

. To me, this is less ideal than the previous version because I didn’t know what to use until I looked at the source code. 😅 ( Maybe there are some documentation I’m missing ) Is this a bug? Happy to create an issue to follow up 🙂

Hi all - quick question about Prisma types and typescript. Getting an error as I'm trying to type the Prisma response object with props as a number, but I get Prisma.Decimal | null is not compatible with number? What can I type this to?

Hello everybody, I cannot set a field

null

if it’s defined as a list ? Like

String[]

?

can anyone help

where's the best way to run a where on a deeply related query? Say if I have the relations A -> B -> C -> D. And I want to return A wth all relations of B, with all relations of C with a relation of D that

type=item

SQL i'm trying to mimic

SELECT *
FROM a
LEFT JOIN b
LEFT JOIN c
LEFT JOIN d
WHERE d.type='item'

this.prisma.A.findMany({
  include: {
    B: {
      include: {
        C: {
          include: {
            d: {
              where: {
                type: 'item'
              }
            }
          }
        }
      }
    }
  }
})

New to Prisma? Are you just starting a new project / startup / side gig? Check out our new free course on how to build a fullstack app using Next.js, GraphQL, TypeScript, and Prisma, hosted by our DevRel Advocate Mahmoud Abdelwahab 🙌 https://www.prisma.io/blog/announcing-upcoming-course-8s41wdqrlgc7

prisma dark I could use your help: We have this small section in our documentation for community generators: https://www.prisma.io/docs/concepts/components/prisma-schema/generators#community-generators Some of them we are even testing in our end to end tests, to make sure everything keeps working: https://github.com/prisma/e2e-tests/tree/dev/community-generators Do you know of any we are missing here?

I would like to change the name and location of the prisma folder. Currently it called

prisma

and it sits on the root folder of my nextjs project. I'd like to name it

db

and have it within the

src

folder. Is that possible? If so, how? Thanks 🙂

Hey alll.... not sure if we've got a weird requirement here....... We are wanting to use Prisma to interface with our database... we've used introspect to generate our schema, and we can deploy that and that works.... However.... the database structure may be updated somewhere else, and deployed separately... meaning that any future deployments of our database interface would possibly break.. I assuming this is where migrate comes in, and we need to be using migrate for updating our database structure now?

hello, I have a raw query (postgres) here with an

ORDER BY

this.prisma.$queryRaw`
	SELECT *
	FROM "public"."OrderStatus"
	INNER JOIN "public"."OrderStatusDescription" ON ("public"."OrderStatus"."id" = "public"."OrderStatusDescription"."orderStatusId")
	WHERE "public"."OrderStatusDescription"."languageId" = ${languageId}
	ORDER BY "public"."OrderStatusDescription"."name" DESC LIMIT ${pagination.take} OFFSET ${pagination.skip}
`

the moment I have a variable instead of hardcoded

DESC

the query doesnt work.

ORDER BY "public"."OrderStatusDescription"."name" ${order.toUpperCase()} // order = 'asc' | 'desc'

Hi everyone ! I’m upgrading to prisma2 but facing an issue. One app usecase involves my api making 3 consecutive queries on the same table (with different WHERE params) in my postgres db. If repeating the operation, 2 or 3 times in a row, one database process blows up the cpu usage. I’ve tried to tune the _connection_limit_ param (setting it to 1) with no luck. Could this be related to the query lock mode ? I’m a bit confused by the pg_locks output. Have you already faced something similar ? Thanks prisma team for the great work !

Hey folks, Would love some help with this non-descriptive error message I'm getting (see screenshot). I call

prisma.query.create

as follows (where queryService(data) is just calling

prisma.create({data})

):

const query = await this.queryService.createQuery({
            genome: s.slice(0,10),
            type: 'Auto',
            description: form.description,
            user: {
                connect: {id: user.id}
            }
        });

This produces the error pretty much no matter what I do. I logged the object I'm trying to insert, seems fine... so I don't see what the problem is. The relevant model is:

model Query {
  id String @id @default(uuid())
  createdAt DateTime @default(now())
  updatedAt DateTime @updatedAt

  user User @relation(fields: [userId], references: [id])
  userId String

  genome String
  type GenomeType @default(Auto)
  description String?
  result String?
}

Would appreciate any help!

I want to modify data from a prisma query before users receive the data through the various

prisma.modelName.queryName

methods - but after the query successfully complets. As a contrived example, let’s say I have a model like this:

model Person {
   id @id @default(cuid())

   first_name String
   
   gender Gender
}

Now what if I want to modify

first_name

after fetching it from the database to include

Mr

or

Ms

. so that if i do:

const bob = await prisma.person.findFirst({
  where: { id: someCuid }
})

Then

bob.first_name

should say

Mr. bob

or something like that. I know that Prisma has middleware using

prisma.$use

but I don’t quite understand how i’d go about doing this since this is a post-query modification … as opposed to a pre-query modification that I see examples of in the docs

hello guys

can i use native db functions like TRIM from pgsql?

like: select field from table where trim(fieldB) = 'whatever '

Hey, I had a question around best practice. So we have a few different services that interact with our DB using prisma. We haven't yet started using

prisma migrate

and currently we use sequelize migrations to make changes to our DB. So whenever we need to make a change to our DB our process is: • Run sequelize migrate • run prisma introspect manually on all our services that use prisma & commit I am just wondering if there's a better way to do this? (eg: Should we be generating our prisma schemas on build time?)