Please, can somebody guide me in creating a Ktor server using graphql-kotlin? If there's a way in the first place?

Hey, I'm trying to inspect the request when creating a

GraphQLContext

to be able to customize a callout to an external service, but I can't find how to get hold of the actual query body from the request? Is there a way to get hold of the raw query when creating my context?

Hi there, does anyone know how I would go about grabbing the requested fields that are from a fragment? I’ve looked into the

DataFetchingEnvironment.selectionSet.definitions

object which seems to be what I want (or at least partially), but it doesn’t appear to contain fields requested from fragments.

i found a way to make a dataloader implementation for graphql-kotlin that allows using suspended functions (based on an approach proposed by apollo) https://gist.github.com/lennyburdette/f3fe6ae7a498698774cc95d1bfc956b4 … would love some review/opinions!

Hi everyone! I am currently working on implementing custom error handling and logging using GraphQl and Kotlin on a federated service. Does anyone have a current example on how to do this? The examples I've found online seem to use deprecated classes.

Hi, has anyone managed to add an Option<T> as a scalar for graphql?

How is the Caching / eTag Support for the Library? If I use Apollo Client on the Frontend, would I benefit from caching?

Hello, thank you for the great library. I am trying to cache some database calls on per request basis. I thought I could use Spring components with @RequestScope, however that doesn’t seem to be supported with Reactive Stack. Do you have any recommendations on what to use instead?

Hi. Does GraphQL kotlin preserve order of certain functions send in one graphql request? I have something like this:

mutation {
  deleteObj1(where){
    affectedRows
  }

  deleteObj2(where){
    affectedRows
  }

  deleteObj3(where){
    affectedRows
  }

  deleteObj4(where){
    affectedRows
  }
}

If order of executed functions will change, there can be thrown foreign key constraint condition -> I need to execute mutations like DELETE CASCADE...

hi folks, jeff from apollo tells me y’all (the expedia team) have figured out a way to check the generated SDL into source control that reflects changes to your kotlin classes — is that true? how does it work?

Hi, We want to introduce a Kotlin Service with GraphQL now. Currently our whole ecosystem is WebMVC. I saw that graphql-kotlin supports WebFlux exclusively at the moment. Is there a plan to change that, eg. to extract functionality to enable others to choose which eco system they wanna use? Many thanks

Hi is there a way to disable adding “Input” suffix for Graphql input types?

Hi, We are now using a compromise to the starter, we use the graphql-kotlin-schema-generator. So far it works well. We use Type Aliases for BigDecimal to make it clearer what it actually is (percent/euro). How do I get these typeAliases as Scalars into the schema? I saw this page explains custom scalars very nicely. BUT I was wondering if it would be possible to reuse the scalar that seems to exist for BigDecimal already? At least the mapping seems to work, but in the schema it says just BigDecimal, even if we have the typealiases in place.

Hello! Is there any way to add Kotlin extension functions to schema generation? I have read some issues related to my question - if I understand correctly, graphql kotlin does not provide special funcionality for generating extension functions. But is there any workaround for this?

data class First(
      val id: Long,
      val name: String
) {
   fun second(env: DataFetchingEnvironment): Second {
      //query second class
   }
}

fun First.another(env: DataFetchingEnvironment): Another {
// I need this function to be seen in graphql schema and resolver
}

so.. we are using a shared project for the data classes, use those from graphql sping server and i wrote a few lines for doing the requests using kotlinx-serialization and ktor-client i would like to also use kotlinx-serialization on the server, is there such a project .. or where can i help to make that happen ?

I’m experimenting with graphq-kotlin and I wonder if it possible to get a response which combines errors and data. Currently when any of my data fetcher fails, the whole response fails and there is no

data

section in the response, only

errors

. Is it possible to get a response which would give me values for fetcher who did not fail and errors for the fetchers which fail? Something like this:

{
    "data": {
        "item": {
            "foo": "foo"
        },
    }
    "errors": [
        {
            "message": "Exception while fetching data (error) : I wonder what happends",
            "locations": [
                {
                    "line": 5,
                    "column": 3
                }
            ],
            "path": [
                "bar"
            ],
            "extensions": {}
        }
    ]
}

Another question - is it possible to specify additional extensions in the response? We are thinking about exposing some additional metadata with the response and the extensions field seems to be the right fit for this. Is there a way to send the extensions with the responses coming from the server?

Hi, We are having a problem implementing a separate data fetcher for a property. This documentation says to use a function instead of a property for that. How would you do this using spring beans in a spring-MVC environment? As this is a POJO still you cannot really inject any repote clients here. Any ideas how we can achieve what we wanna do?

Hey guys, I have a question regarding requesting non-existing fields. Currently when I ask for a field which does not exist in the schema, the whole query fails, but I wonder if it would be somehow possible to return the other fields which are valid and return errors (in the errors field) only for the fields which do not exists in the schema.

Is there any recommendation when it comes to publishing the schema? I’m running kotlin-graphql in spring and I’m wondering what would be the best way to integrate the schema exposed by this service with Apollo Studio? Pretty much what I’m looking for is an ability to generate the schema to a file and then upload it to Apollo Studio. Ideally all this happening in our CI build ..

I’m investigating porting our existing Kotlin graphql server to graphql-kotlin. One thing I’m not quite clear on from docs — I see that you specify packages to scan for classes, but how does that relate to providing root types? eg, if a type in one of the packages references (arguments, return type, etc) a type from a faraway package, will that type be processed into the schema too, or not? and do all types in the specified packages get treated as graphql types by default, or only those reachable from the root types?

Is there a way to use generic types with the schema generator? What I’m looking for is being able to define an abstract class with a generic type and make it “excluded” from the schema, but still expose the functions defined by the interface in the schema together with the resolved type at the use time. Something like this:

abstract class Page<T:Any> {
   @GraphQLIgnore
   abstract fetchItems() ... 

   // these will be exposed in the schema
   fun itemsCount(): Int
   fun hasNextPage(): Boolean
   fun pageItems: List<T>
}

class PageOfStrings: Page<String> {
   override fetchItems() ... 
}

class PageOfInts Page<Int> {
   override fetchItems() ... 
}

The resulting schema would have two types:

type PageOfStrings {
   itemsCount: Int!
   hasNextPage: Boolean!
   pageItems: [String]!
}

type PageOfInts {
   itemsCount: Int!
   hasNextPage: Boolean!
   pageItems: [Int]!
}

Is something like this doable?

Hi! Trying graphql-kotlin for the first time with 3.6.8 and 4.0.0-alpha.7 for a small control plane, and running into trouble with:

@JsonTypeInfo(use = JsonTypeInfo.Id.NAME)
sealed class Command {
	@JsonTypeName("ResetCursorPosition")
	data class ResetCursorPosition(
		val dataSource: String, val position: Instant
	) : Command()
}

which causes a failure in

assertTypeUniqueness

; might I be missing some semi-obvious gotchas with the current implementation? Is it worth trying to minimize to a reproduction case? Otherwise the DX has been great so far.

Hi. I’m trying to add a conversion from Set<*> to a GraphQLList. I know it is not inherently supported in GQL but the backing data is read-only and has already been implemented using sets. For simplicity lets say we have:

data class TestClass(
   val content: Set<String>
)

to remedy this, I’ve added a Hook with the following SchemaGeneratorHooks#willGenerateGrpahQLType method:

@ExperimentalStdlibApi
override fun willGenerateGraphQLType(type: KType): GraphQLType? = when {
    type.isSubtypeOf(typeOf<Set<*>?>()) -> {
        type.arguments.first().type?.let { kType ->
            val typeRef = (kType.classifier as KClass<*>)
            GraphQLList.list(GraphQLTypeReference.typeRef(typeRef.simpleName))
        }
    }
    else -> null
}

this works when creating a query that returns TestClass or List<TestClass>. However, when creating a Query that returns a Set<TestClass> it fails as the TestClass reference has not yet been created. From what I could gather, this method should just create the reference to

TestClass::class.simpleName

and the generator should generate the class in “due-time” which should lead to Set<TestClass> resolving as a GQLList<TestClass> in the playground, or am I missing something?

I am trying to accomplish authentication with JWTs. here: https://github.com/ExpediaGroup/graphql-kotlin/discussions/663In it was advised to do some JWT validation work in a spring webfilter and then store that is the security context and pick that up in the GraphQl context. Is this totally necessary or could all the JWT validation logic just be dont in the GraphQLContextFactory generateContext function, as we have access to headers there?

Hey, there! I'm trying to incorporate graphql-kotlin into an existing spring boot rest application with starter-security, starter-web, etc. I'm wondering if there's any special configuration or caveats I should know about? I've created a few root-level queries, and the schema generation works as expected, but the /graphql endpoint isn't working (404), nor is the playground.

him we are trying to add subscription (now that we got websockets to go through the google cloudrun) and it seems we are running into a few weird issues whenever there is any issues (syntax or such) it returns

{
  "error": "Could not connect to websocket endpoint <ws://localhost:8080/subscriptions>. Please check if the endpoint url is correct."
}

not sure if this is a issue in plaground or in the library however and seems like the

GraphQLContextFactory

does not kick in for subscriptions so we cannot figure out how to get access to Header

Authorization

for subscriptions currently

Hey, I'm trying to register a

FederatedTypeResolver

in a Boot applications, but how ever I do it I can't get it to kick in. I wonder if I'm misunderstanding somehing - I have a simple schema like this

@KeyDirective(FieldSet("id"))
data class MySubThing(val id: Int?, val extraThing: String)

data class MyThing(@ExtendsDirective val id: String, val subThings: List<MySubThing>) {
}

, and register this resolver class as a bean:

class MySubThingResolver: FederatedTypeResolver<MySubThing> {
    override suspend fun resolve(
      environment: DataFetchingEnvironment,
      representations: List<Map<String, Any>>
    ): List<MySubThing?> {
      return representations.stream().map {
        MySubThing(it["id"]?.toString()?.toInt(), "WOOOP")
      }.collect(Collectors.toList())
    }

But I can't get the resolver to be called however I try. What is the recommended way of registering resolvers for a schema in a Boot app? (I have set the

graphql.federation.enabled

to

true

)

This page (https://expediagroup.github.io/graphql-kotlin/docs/schema-generator/writing-schemas/interfaces) suggests it should be possible to use sealed classes in our generated schema. However, I’ve not been able to make this work for us. Is this known broken in the current 4.0 alpa11 or is there something I’m missing here? I’m trying to add a field to one of our input types that is a simple sealed class. As soon as I do that, I get this error:

Caused by: com.expediagroup.graphql.exceptions.CouldNotCastGraphQLSchemaElement: Could not cast GraphQLSchemaElement GraphQLInterfaceType{name='GeoJsonGeometry', description='null', fieldDefinitionsByName=[type], typeResolver=null} to class graphql.schema.GraphQLInputType

This is the sealed class

sealed class GeoJsonGeometry(val type: GeometryType) {
    class Point(type: GeometryType = GeometryType.Point): GeoJsonGeometry(type)
    class Polygon(type: GeometryType = GeometryType.Polygon): GeoJsonGeometry(type)
}

Simply adding a

val geometry: GeoJsonGeometry?=null

to our input date class breaks it, removing that fixes it.

Any work (or even just thinking) been done on subscription protocol migration from https://github.com/apollographql/subscriptions-transport-ws to https://github.com/enisdenjo/graphql-ws ?