type User { friends: [User!]! } <= datamodel.graphql Does that field have some issues when it comes to friends_some? I don't understand what is scalar and not xD

is there a way in graphql to query just the most recently created object of a given type...without knowing anything about that item? Just "there are things and give me the most recent thing that was added"?

Everything seems to want a

where

selector to be passed, and that seems separate from sorting

Has anyone migrated from graphcool to prisma? Any good tutorials or docs on the process? I was using graphcool cause I have no real backend experience

Started having issues with

graphql codegen

I updated the model with extra types and it started to give this error. Even when reverting to a model that did work still gives this error. Anyone came across this?

{ GraphQLError: Syntax Error: Expected Name, found }
    at syntaxError (/Users/.../projects/.../node_modules/graphql/error/syntaxError.js:24:10)
    at expect (/Users/.../projects/.../node_modules/graphql/language/parser.js:1299:32)
    at parseName (/Users/.../projects/.../node_modules/graphql/language/parser.js:92:15)
    at parseFieldDefinition (/Users/.../projects/.../node_modules/graphql/language/parser.js:809:14)
    at many (/Users/.../projects/.../node_modules/graphql/language/parser.js:1348:16)
    at parseFieldsDefinition (/Users/.../projects/.../node_modules/graphql/language/parser.js:799:50)
    at parseObjectTypeDefinition (/Users/.../projects/.../node_modules/graphql/language/parser.js:757:16)
    at parseTypeSystemDefinition (/Users/.../projects/.../node_modules/graphql/language/parser.js:663:16)
    at parseDefinition (/Users/.../projects/.../node_modules/graphql/language/parser.js:143:16)
    at parseDocument (/Users/.../projects/.../node_modules/graphql/language/parser.js:110:22)
  message: 'Syntax Error: Expected Name, found }',
  locations: [ { line: 193, column: 1 } ],
  path: undefined }

I think I know why this error occurs. It comes up not on codegen but on

graphql get-schema --project database

and only when there is only a collection element.

type Team {
  members: [Participant!]!
}

This will cause the error, but can be corrected if

name: String

(or the like) is added. But only after deleting the generated files. It may be down to the lack of unique ID also (which it should have)

is the old Graphcool slowly being deprecated? I noticed I can no longer see function logs

Hello! At the end of this blog post https://blog.graph.cool/how-to-deploy-a-prisma-cluster-to-aws-fargate-using-docker-cloudformation-293aa8727b89, it says:

If you don’t want to fiddle with CloudFormation and the AWS GUI yourself, you can also use Prisma Cloud to easily spin up your own Prisma Clusters.

but I couldn’t find any information on how to create a Private cluster/server in Prisma Cloud. Is that available on the paid plan only? Thanks!

I have a Prisma Fargate instance running and it works great. One thing I'm wondering is why there are two VPCs. Would a PR that put the db and the server into one VPC, with public/private subnets be appreciated or is there something I'm missing here?

Is anyone here using GraphQL-Yoga on AWS Lambda? I’ve sort of gotten it running, but I’m getting this error:

Warning: handler 'graphql' returned a promise and also use a callback! This is problematic and might cause issues in you lambda.

I’m following this example mostly: https://github.com/prismagraphql/graphql-yoga/blob/master/examples/lambda/serverless.yml

hey @matic so I was trying to make a directives version of GraphQL Shield but I got sidetracked with what I really have to do, so I decided to give GraphQL Shield a try and I'v got to say, I LOVE IT! Logic-based rules are so flexible, it is a really good middleware you made there... Anyway, I wrote a rules class with useful methods for a RBAC-type of permissions, though I'm not quite sure if it sticks to the RBAC definition at all, but it has roles and scopes and authentication check, I thought you might want to give it a look and see if it makes any sense at all: https://gist.github.com/dukuo/351f581e54b0af863f7155110e8f39c6 . looking forward to any comments from @everyone btw! Next step is to write unit tests to properly check that everything's in order.

This message was deleted.

Can anyone from devs handle this isuue? It is very annoying https://github.com/prismagraphql/prisma-binding/issues/182

Hi All, How can I export data from existing GraphCool project? I don't have any setup files for a project locally yet

Hello everyone prisma green ! I'm new to GraphQL and happy to start learning it with Prisma !

Hey guys, I was wondering for the most part is seems like Prisma allows you to run queries/mutations on the database. But I was wondering how are cases handled where you want to run a mutation that doesn't directly alter database records (e.g. you need to communicate with an external API or run some background task, etc). Thank you!

After reading the API docs about nested mutation (https://www.prisma.io/docs/reference/prisma-api/mutations-ol0yuoz6go#nested-mutations), i'm a little lost on how I could write resolvers for nested mutation in the application layer... Does anyone have learning resource(s) for this ? 🤔

is there an easy way to take a schema defined in SDL and convert it to a .js file? I know the graphql-cli can do it, but seems like only via an introspection query to a running server. Any way to do it based on just the SDL?

what the ad

Please don't post spam, thanks 🙂

it's banhammer time

Hi - I just got prisma running on GCE/kubernetes, and I have an exposed endpoint. When I try to prisma add cluster I get an error about this command being deprecated and unavailable in prisma 1.9 (which is the version I’m using), and when I try to prisma deploy to the endpoint, I get an error about the PRISMA_MANAGEMENT_API_SECRET, which is not in the deployment yml in the kubernetes tutorial…

I really just want to have a backend service for prisma, I’m not married to GCE, so if there’s a simpler/easier way I’m all for it. My understanding is that Prisma Cloud isn’t ready for production projects yet

hi friends i can cache graphql query in client side ? what's the best method ?

I am confused with how a subscription endpoint is exposed via

graphql-yoga

I have the following options config:

const options = {
    port: PORT,
    cors: {
      credentials: true,
      origin:
        process.env.NODE_ENV === 'production' && !process.env.FORCE_DEV
          ? ['<https://xxx.app>', /xxx-(\w|-)+\.now\.sh/g, /xxx\.app/]
          : [/localhost/]
    },
    endpoint: '/graphql',
    subscriptions: '/subscriptions',
    playground: '/playground'
  }

Oh Snap! Look what just popped up in my Podcast Feed: https://devchat.tv/js-jabber/jsj-317-prisma-with-johannes-schickling cc @schickling

Hey guys anybody knows if i can still book the regular ticket for graphQL-europe event

Hey guys, I’m a bit stuck on something and I don’t know if I’m missing something in the docs or if I’m just confused. I have a list of locations that need to be populated from the server, and only want to get the first unique result from each node. Basically I want to do the equivalent of “DISTINCT” in SQL. How would I approach this in Prisma? EDIT: Found this: https://www.prisma.io/forum/t/how-to-select-distinct/3687 I guess it’s not possible yet.

Quick Q: Is there a way to set a @unique field or any field to null in a mutation?