F# Support

[michaelstaib]

This PR represents the initial work to determine what we need to make this a great API for F#.

In this initial PR we will re-create the StarWars example with F# and write some F# unit tests to see where the pain points are.

Todo:

• [ ] F# StarWars
• [ ] F# Integration Tests
• [ ] F# Union Support

https://github.com/adelarsq/hotchocolate_fsharp_sample

[sep2]

Is it possible to represent non-nullable GraphQL type by F# record? now I can do nullable type by this method:

[<ReferenceEquality>]
type Author = { Name: string }

[<ReferenceEquality>]
type Book = { Title: string; Author: Author }

type Query() =
    member this.Book() =
        { Title = "F#"
          Author = { Name = "Jon" } }

type Mutation() =
    member this.CreateBook() =
        { Title = "F#"
          Author = { Name = "Jon" } }

[sep2]

Never mind, I found this [<GraphQLNonNullType>] attribute is exactly what I want.

[sonarcloud[bot]]

Kudos, SonarCloud Quality Gate passed!

0 Bugs
0 Vulnerabilities
0 Security Hotspots
0 Code Smells

No Coverage information
No Duplication information

[Zaid-Ajaj]

Hi @michaelstaib F# support with HotChocolate would be awesome! I will be looking forward to testing it out with Snowflaqe 🙏

[geekox86]

Hello @michaelstaib. Are we expecting this in Sept. milestone? What would it include in terms of F# support?

[Ciantic]

This is a bit cumbersome in F#:

type Query() =
    member this.Book() =
        { Title = "F#"
          Author = { Name = "Jon" } }

type Mutation() =
    member this.CreateBook() =
        { Title = "F#"
          Author = { Name = "Jon" } }

There is no partial classes, it gets awkward. Secondly these classes feel redundant, especially in F#. Mutation functions usually are not dependent on each other by state, so they shouldn't share class anyway.

I'd prefer in F# if we could define mutations and queries like functions. If it's possible, by allowing somehow to annotate functions with "Mutation":

[<Mutation>]
let createBook book (appDbContext: AppDbContext) =
    appDbContext.Books.Add(book) |> ignore
    appDbContext.SaveChanges()
    // ... 

[<Query>]
[<UsePaging>]
[<UseProjection>]
[<UseFiltering>]
[<UseSorting>]
let getBooks ([<Service>] dbContext: AppDbContext) =
    dbContext.Books

[michaelstaib]

@Ciantic are there global functions in F# or are they always members of a class or struct?

[michaelstaib]

OK, looked at it ... seems to work ... I can probably put something into one of the next builds ... I think with this kind of function it can also become quite nice to create schema first and define resolvers... have to play a bit with it

[Ciantic]

It's notable that in F# the functions are usually defined in curried form, which also requires support in the HotChocolate to work well. Like input -> AppDbContext -> PersonPayload I also started my simple example repo here:

https://github.com/Ciantic/TestGraphQLWithFsharp/blob/master/Program.fs

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[mika76]

I'm sure there's plenty of people who wouldn't want this closed?

[adelarsq]

I am interested on this too.

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[ss-proc]

@michaelstaib Please mark this as not stale

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you all sign our Contributor License Agreement before we can accept your contribution.
1 out of 2 committers have signed the CLA.

:white_check_mark: michaelstaib
:x: wonbyte
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[mika76]

was this PR forgotten?

[michaelstaib]

I am closing this PR as we have a newer one #5595.