😁hey everybody did upgraded graphql-yoga to graphql-server 2.0 (version 2) ?

I'm trying to run some fairly straight forward tests against my graphql backend, but though queries work, mutations fail with

Project not found: 'backend$test'

- this is a snippet of the code (again, queries work just fine) https://git.io/fhzoX - anyone seen this kind of error (or know of good testing patterns - without mocks)

huh, actually, queries don't work - they fail when they try to hit the database with the same error.

I can't seem to get

items

to load with this query:

course(where: {id: "cjqqp8g9z00xx0a06kchfp119"}) {
    id
    name
    units {
      id
      name
      items {
        id
        name
      }
    }
  }
}

it loads them in the introspection playground but not my api one. In my api I'm using

forwardTo

Are local deployments also rate limited?

Has anyone in here had a problem with subscriptions? I’m using react + apollo and I’m receiving updates on my frontend about 50% of the time. I’ve been messing around with it for a while and can’t seem to get it 100% working.

Got a bit of noob question: I'm running mutations that add nested data via

{ create: { ... }

or

{ connect: { ... } }

and I can see the records in my postgres database (in the tables starting with _), but when I query across data relations, it never comes back - nothing more complicated than this example: https://www.prisma.io/docs/prisma-graphql-api/reference/queries-qwe1/#querying-data-across-relations

Question being: could I be missing something obvious?

Like do I need to turn something on for Prisma to resolve queries across tables?

Hello. Does

graphqlgen

support `union`s?

👋

hey, I am trying to figure out a way to do something... not sure if it is possible

basically if I have an object like

Post

, and it has a sub object like

Comment

... and

Comment

has an

id

and I want to add some data to

Comment

in the resolver where the data doesn't come from prisma

I can fetch the data from my other source by the

id

of the comment, in my comment resolver

the problem is the

id

is only there if the user queries for

id

in the original query, if they don't query for

id

I don't know how to look up the rest of the data..

basically in my comment resolver, if the user originally queried for id, the id is there and I can add the other props, but if not, the id is undefined and I have no way of looking up the other props... and no way to know which

Comment

they are referencing to go back to prisma and get the

id

hard to explain what I am asking, trying to figure out how to better word it...

the only thing I can figure is basically starting at the top of the resolver chain and completely ignoring their query and fetching everything... but that seems very imperformant

@CCBCodeMonkey not sure I can help, but if you're able to share your resolver for the Comment type, it might help?

I kind of made of that scenario because my schema and resolvers are too complext to post lol

trying to break it down to the simplest way I can explain

im just saying, I have a parent resolver, and a child resolver, and in the child resolver I want to add some props that come from outside prisma, but I have no way of looking them up unless the parent resolver passes in some sort of ID to the child resolver

like I have no reference to the prisma object, just the fields it returned

Does

prisma.user(…)

do any caching? I'm starting to add relation resolvers in my

User

type, but I'm repeating

prisma.user(…).<relation>

throughout my

/Types/User.js

resolver. Just wary that if there's no caching, it could be running the same query against the user table many times…

hey friends, I’m hoping for some good advice on nested queries. Supposing I have a query

getClientUser( data: clientUserRetrievalArgs): ResponsePayload

where ResponsePayload is defined as ‘{ token, clientUser }’. token is not represented in the data store, clientUser is. Any advice on how to ensure that nested queries of clientUser data from info are correctly passed?

Hi, I got to an issue with after introspecting existing Postgres DB. (https://github.com/prisma/prisma/issues/2466)

# Type 'ARRAY' is not yet supported.

I realise it is in the feature request, and it is not listed in roadmap. Dose it mean there would not be any update soon(couple of months)? Thanks in advance.

How to modularize schema.graphql definition and still be able to feed it to

graphqlgen

?

I'm trying to implement an invite-system, where each invite has an expiration-date, and a status that tells me if the invite is still pending, or already expired. right now, I'm computing the value for the status field in my custom resolver based on the expiration-date, which works fine, but I'd rather like it to update the status in the prisma backend based on a timer. how'd I go about doing this? or is this maybe not recommended?

@serum I did something similar but not as exact. Basically I used a serverless function to call a query endpoint and then update the needed records based on the results of the query. Obviously this ended up meaning some could be exipred for a little while waiting for the function to run. Why not just add an expiry date in the backend?